examples/l2fwd-ivshmem: import sample application
authorBruce Richardson <bruce.richardson@intel.com>
Tue, 11 Feb 2014 11:31:03 +0000 (11:31 +0000)
committerDavid Marchand <david.marchand@6wind.com>
Wed, 26 Feb 2014 09:47:59 +0000 (10:47 +0100)
This provides a new sample application which demonstrates how
the ivshmem library and EAL capabilities can be used to create
a zero-copy fast-path for packet communication between host
machine and guest vm.

Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>
examples/l2fwd-ivshmem/Makefile [new file with mode: 0644]
examples/l2fwd-ivshmem/guest/Makefile [new file with mode: 0644]
examples/l2fwd-ivshmem/guest/guest.c [new file with mode: 0644]
examples/l2fwd-ivshmem/host/Makefile [new file with mode: 0644]
examples/l2fwd-ivshmem/host/host.c [new file with mode: 0644]
examples/l2fwd-ivshmem/include/common.h [new file with mode: 0644]

diff --git a/examples/l2fwd-ivshmem/Makefile b/examples/l2fwd-ivshmem/Makefile
new file mode 100644 (file)
index 0000000..7286b37
--- /dev/null
@@ -0,0 +1,50 @@
+#   BSD LICENSE
+# 
+#   Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
+#   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 Intel Corporation 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.
+
+ifeq ($(RTE_SDK),)
+$(error "Please define RTE_SDK environment variable")
+endif
+
+# Default target, can be overriden by command line or environment
+RTE_TARGET ?= x86_64-ivshmem-linuxapp-gcc
+
+include $(RTE_SDK)/mk/rte.vars.mk
+unexport RTE_SRCDIR RTE_OUTPUT RTE_EXTMK
+
+DIRS-$(CONFIG_RTE_EXEC_ENV_LINUXAPP) += host guest
+
+.PHONY: all clean $(DIRS-y)
+
+all: $(DIRS-y)
+clean: $(DIRS-y)
+
+$(DIRS-y):
+       $(MAKE) -C $@ $(MAKECMDGOALS)
diff --git a/examples/l2fwd-ivshmem/guest/Makefile b/examples/l2fwd-ivshmem/guest/Makefile
new file mode 100644 (file)
index 0000000..04e13d7
--- /dev/null
@@ -0,0 +1,50 @@
+#   BSD LICENSE
+# 
+#   Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
+#   All rights reserved.
+# 
+#   Redistribution and use in source and binary forms, with or without
+#   modification, are permitted provided that the following conditions
+#   are met:
+# 
+#     * Redistributions of source code must retain the above copyright
+#       notice, this list of conditions and the following disclaimer.
+#     * Redistributions in binary form must reproduce the above copyright
+#       notice, this list of conditions and the following disclaimer in
+#       the documentation and/or other materials provided with the
+#       distribution.
+#     * Neither the name of Intel Corporation nor the names of its
+#       contributors may be used to endorse or promote products derived
+#       from this software without specific prior written permission.
+# 
+#   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+#   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+#   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+#   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+#   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+#   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+#   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+#   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+#   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+#   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+#   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+ifeq ($(RTE_SDK),)
+$(error "Please define RTE_SDK environment variable")
+endif
+
+# Default target, can be overriden by command line or environment
+RTE_TARGET ?= x86_64-ivshmem-linuxapp-gcc
+
+include $(RTE_SDK)/mk/rte.vars.mk
+
+# binary name
+APP = guest
+
+# all source are stored in SRCS-y
+SRCS-y := guest.c
+
+CFLAGS += -O3
+CFLAGS += $(WERROR_FLAGS)
+
+include $(RTE_SDK)/mk/rte.extapp.mk
diff --git a/examples/l2fwd-ivshmem/guest/guest.c b/examples/l2fwd-ivshmem/guest/guest.c
new file mode 100644 (file)
index 0000000..adcbe79
--- /dev/null
@@ -0,0 +1,453 @@
+/*-
+ *   BSD LICENSE
+ * 
+ *   Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
+ *   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 Intel Corporation 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 <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <stdint.h>
+#include <unistd.h>
+#include <getopt.h>
+#include <signal.h>
+#include <sys/mman.h>
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <sys/queue.h>
+#include <sys/file.h>
+#include <unistd.h>
+#include <limits.h>
+#include <errno.h>
+#include <sys/ioctl.h>
+#include <sys/time.h>
+
+#include <rte_common.h>
+#include <rte_eal_memconfig.h>
+#include <rte_log.h>
+#include <rte_memory.h>
+#include <rte_memcpy.h>
+#include <rte_memzone.h>
+#include <rte_tailq.h>
+#include <rte_eal.h>
+#include <rte_per_lcore.h>
+#include <rte_launch.h>
+#include <rte_atomic.h>
+#include <rte_cycles.h>
+#include <rte_prefetch.h>
+#include <rte_lcore.h>
+#include <rte_per_lcore.h>
+#include <rte_branch_prediction.h>
+#include <rte_interrupts.h>
+#include <rte_pci.h>
+#include <rte_random.h>
+#include <rte_debug.h>
+#include <rte_ether.h>
+#include <rte_ethdev.h>
+#include <rte_ring.h>
+#include <rte_mempool.h>
+#include <rte_mbuf.h>
+#include <rte_ivshmem.h>
+
+#include "../include/common.h"
+
+#define MAX_RX_QUEUE_PER_LCORE 16
+#define MAX_TX_QUEUE_PER_PORT 16
+struct lcore_queue_conf {
+       unsigned n_rx_port;
+       unsigned rx_port_list[MAX_RX_QUEUE_PER_LCORE];
+       struct mbuf_table rx_mbufs[RTE_MAX_ETHPORTS];
+       struct vm_port_param * port_param[MAX_RX_QUEUE_PER_LCORE];
+} __rte_cache_aligned;
+static struct lcore_queue_conf lcore_queue_conf[RTE_MAX_LCORE];
+
+/* Print out statistics on packets dropped */
+static void
+print_stats(void)
+{
+       uint64_t total_packets_dropped, total_packets_tx, total_packets_rx;
+       unsigned portid;
+
+       total_packets_dropped = 0;
+       total_packets_tx = 0;
+       total_packets_rx = 0;
+
+       const char clr[] = { 27, '[', '2', 'J', '\0' };
+       const char topLeft[] = { 27, '[', '1', ';', '1', 'H','\0' };
+
+               /* Clear screen and move to top left */
+       printf("%s%s", clr, topLeft);
+
+       printf("\nPort statistics ====================================");
+
+       for (portid = 0; portid < ctrl->nb_ports; portid++) {
+               /* skip ports that are not enabled */
+               printf("\nStatistics for port %u ------------------------------"
+                          "\nPackets sent: %24"PRIu64
+                          "\nPackets received: %20"PRIu64
+                          "\nPackets dropped: %21"PRIu64,
+                          portid,
+                          ctrl->vm_ports[portid].stats.tx,
+                          ctrl->vm_ports[portid].stats.rx,
+                          ctrl->vm_ports[portid].stats.dropped);
+
+               total_packets_dropped += ctrl->vm_ports[portid].stats.dropped;
+               total_packets_tx += ctrl->vm_ports[portid].stats.tx;
+               total_packets_rx += ctrl->vm_ports[portid].stats.rx;
+       }
+       printf("\nAggregate statistics ==============================="
+                  "\nTotal packets sent: %18"PRIu64
+                  "\nTotal packets received: %14"PRIu64
+                  "\nTotal packets dropped: %15"PRIu64,
+                  total_packets_tx,
+                  total_packets_rx,
+                  total_packets_dropped);
+       printf("\n====================================================\n");
+}
+
+/* display usage */
+static void
+l2fwd_ivshmem_usage(const char *prgname)
+{
+       printf("%s [EAL options] -- [-q NQ -T PERIOD]\n"
+                  "  -q NQ: number of queue (=ports) per lcore (default is 1)\n"
+                  "  -T PERIOD: statistics will be refreshed each PERIOD seconds (0 to disable, 10 default, 86400 maximum)\n",
+              prgname);
+}
+
+static unsigned int
+l2fwd_ivshmem_parse_nqueue(const char *q_arg)
+{
+       char *end = NULL;
+       unsigned long n;
+
+       /* parse hexadecimal string */
+       n = strtoul(q_arg, &end, 10);
+       if ((q_arg[0] == '\0') || (end == NULL) || (*end != '\0'))
+               return 0;
+       if (n == 0)
+               return 0;
+       if (n >= MAX_RX_QUEUE_PER_LCORE)
+               return 0;
+
+       return n;
+}
+
+static int
+l2fwd_ivshmem_parse_timer_period(const char *q_arg)
+{
+       char *end = NULL;
+       int n;
+
+       /* parse number string */
+       n = strtol(q_arg, &end, 10);
+       if ((q_arg[0] == '\0') || (end == NULL) || (*end != '\0'))
+               return -1;
+       if (n >= MAX_TIMER_PERIOD)
+               return -1;
+
+       return n;
+}
+
+/* Parse the argument given in the command line of the application */
+static int
+l2fwd_ivshmem_parse_args(int argc, char **argv)
+{
+       int opt, ret;
+       char **argvopt;
+       int option_index;
+       char *prgname = argv[0];
+       static struct option lgopts[] = {
+               {NULL, 0, 0, 0}
+       };
+
+       argvopt = argv;
+
+       while ((opt = getopt_long(argc, argvopt, "q:p:T:",
+                                 lgopts, &option_index)) != EOF) {
+
+               switch (opt) {
+
+               /* nqueue */
+               case 'q':
+                       l2fwd_ivshmem_rx_queue_per_lcore = l2fwd_ivshmem_parse_nqueue(optarg);
+                       if (l2fwd_ivshmem_rx_queue_per_lcore == 0) {
+                               printf("invalid queue number\n");
+                               l2fwd_ivshmem_usage(prgname);
+                               return -1;
+                       }
+                       break;
+
+               /* timer period */
+               case 'T':
+                       timer_period = l2fwd_ivshmem_parse_timer_period(optarg) * 1000 * TIMER_MILLISECOND;
+                       if (timer_period < 0) {
+                               printf("invalid timer period\n");
+                               l2fwd_ivshmem_usage(prgname);
+                               return -1;
+                       }
+                       break;
+
+               /* long options */
+               case 0:
+                       l2fwd_ivshmem_usage(prgname);
+                       return -1;
+
+               default:
+                       l2fwd_ivshmem_usage(prgname);
+                       return -1;
+               }
+       }
+
+       if (optind >= 0)
+               argv[optind-1] = prgname;
+
+       ret = optind-1;
+       optind = 0; /* reset getopt lib */
+       return ret;
+}
+
+/*
+ * this loop is getting packets from RX rings of each port, and puts them
+ * into TX rings of destination ports.
+ */
+static void
+fwd_loop(void)
+{
+
+       struct rte_mbuf *pkts_burst[MAX_PKT_BURST];
+       struct rte_mbuf **m_table;
+       struct rte_mbuf *m;
+       struct rte_ring *rx, *tx;
+       unsigned lcore_id, len;
+       uint64_t prev_tsc, diff_tsc, cur_tsc, timer_tsc;
+       unsigned i, j, portid, nb_rx;
+       struct lcore_queue_conf *qconf;
+       struct ether_hdr *eth;
+       void *tmp;
+
+       prev_tsc = 0;
+       timer_tsc = 0;
+
+       lcore_id = rte_lcore_id();
+       qconf = &lcore_queue_conf[lcore_id];
+
+       if (qconf->n_rx_port == 0) {
+               RTE_LOG(INFO, L2FWD_IVSHMEM, "lcore %u has nothing to do\n", lcore_id);
+               return;
+       }
+
+       RTE_LOG(INFO, L2FWD_IVSHMEM, "entering main loop on lcore %u\n", lcore_id);
+
+       for (i = 0; i < qconf->n_rx_port; i++) {
+               portid = qconf->rx_port_list[i];
+               RTE_LOG(INFO, L2FWD_IVSHMEM, " -- lcoreid=%u portid=%u\n", lcore_id,
+                       portid);
+       }
+
+       while (ctrl->state == STATE_FWD) {
+               cur_tsc = rte_rdtsc();
+
+               diff_tsc = cur_tsc - prev_tsc;
+
+               /*
+                * Read packet from RX queues and send it to TX queues
+                */
+               for (i = 0; i < qconf->n_rx_port; i++) {
+
+                       portid = qconf->rx_port_list[i];
+
+                       len = qconf->rx_mbufs[portid].len;
+
+                       rx = ctrl->vm_ports[portid].rx_ring;
+                       tx = ctrl->vm_ports[portid].dst->tx_ring;
+
+                       m_table = qconf->rx_mbufs[portid].m_table;
+
+                       /* if we have something in the queue, try and transmit it down */
+                       if (len != 0) {
+
+                               /* if we succeed in sending the packets down, mark queue as free */
+                               if (rte_ring_enqueue_bulk(tx, (void**) m_table, len) == 0) {
+                                       ctrl->vm_ports[portid].stats.tx += len;
+                                       qconf->rx_mbufs[portid].len = 0;
+                                       len = 0;
+                               }
+                       }
+
+                       nb_rx = rte_ring_count(rx);
+
+                       nb_rx = RTE_MIN(nb_rx, (unsigned) MAX_PKT_BURST);
+
+                       if (nb_rx == 0)
+                               continue;
+
+                       /* if we can get packets into the m_table */
+                       if (nb_rx < (RTE_DIM(qconf->rx_mbufs[portid].m_table) - len)) {
+
+                               /* this situation cannot exist, so if we fail to dequeue, that
+                                * means something went horribly wrong, hence the failure. */
+                               if (rte_ring_dequeue_bulk(rx, (void**) pkts_burst, nb_rx) < 0) {
+                                       ctrl->state = STATE_FAIL;
+                                       return;
+                               }
+
+                               ctrl->vm_ports[portid].stats.rx += nb_rx;
+
+                               /* put packets into the queue */
+                               for (j = 0; j < nb_rx; j++) {
+                                       m = pkts_burst[j];
+
+                                       rte_prefetch0(rte_pktmbuf_mtod(m, void *));
+
+                                       m_table[len + j] = m;
+
+                                       eth = rte_pktmbuf_mtod(m, struct ether_hdr *);
+
+                                       /* 02:00:00:00:00:xx */
+                                       tmp = &eth->d_addr.addr_bytes[0];
+                                       *((uint64_t *)tmp) = 0x000000000002 + ((uint64_t)portid << 40);
+
+                                       /* src addr */
+                                       ether_addr_copy(&ctrl->vm_ports[portid].dst->ethaddr,
+                                                       &eth->s_addr);
+                               }
+                               qconf->rx_mbufs[portid].len += nb_rx;
+
+                       }
+
+               }
+
+               /* if timer is enabled */
+               if (timer_period > 0) {
+
+                       /* advance the timer */
+                       timer_tsc += diff_tsc;
+
+                       /* if timer has reached its timeout */
+                       if (unlikely(timer_tsc >= (uint64_t) timer_period)) {
+
+                               /* do this only on master core */
+                               if (lcore_id == rte_get_master_lcore()) {
+                                       print_stats();
+                                       /* reset the timer */
+                                       timer_tsc = 0;
+                               }
+                       }
+               }
+
+               prev_tsc = cur_tsc;
+       }
+}
+
+static int
+l2fwd_ivshmem_launch_one_lcore(__attribute__((unused)) void *dummy)
+{
+       fwd_loop();
+       return 0;
+}
+
+int
+main(int argc, char **argv)
+{
+       struct lcore_queue_conf *qconf;
+       const struct rte_memzone * mz;
+       int ret;
+       uint8_t portid;
+       unsigned rx_lcore_id, lcore_id;
+
+       /* init EAL */
+       ret = rte_eal_init(argc, argv);
+       if (ret < 0)
+               rte_exit(EXIT_FAILURE, "Invalid EAL arguments\n");
+       argc -= ret;
+       argv += ret;
+
+       /* parse application arguments (after the EAL ones) */
+       ret = l2fwd_ivshmem_parse_args(argc, argv);
+       if (ret < 0)
+               rte_exit(EXIT_FAILURE, "Invalid l2fwd-ivshmem arguments\n");
+
+       /* find control structure */
+       mz = rte_memzone_lookup(CTRL_MZ_NAME);
+       if (mz == NULL)
+               rte_exit(EXIT_FAILURE, "Cannot find control memzone\n");
+
+       ctrl = (struct ivshmem_ctrl*) mz->addr;
+
+       /* lock the ctrl so that we don't have conflicts with anything else */
+       rte_spinlock_lock(&ctrl->lock);
+
+       if (ctrl->state == STATE_FWD)
+               rte_exit(EXIT_FAILURE, "Forwarding already started!\n");
+
+       rx_lcore_id = 0;
+       qconf = NULL;
+
+       /* Initialize the port/queue configuration of each logical core */
+       for (portid = 0; portid < ctrl->nb_ports; portid++) {
+
+               /* get the lcore_id for this port */
+               while (rte_lcore_is_enabled(rx_lcore_id) == 0 ||
+                          lcore_queue_conf[rx_lcore_id].n_rx_port ==
+                          l2fwd_ivshmem_rx_queue_per_lcore) {
+                       rx_lcore_id++;
+                       if (rx_lcore_id >= RTE_MAX_LCORE)
+                               rte_exit(EXIT_FAILURE, "Not enough cores\n");
+               }
+
+               if (qconf != &lcore_queue_conf[rx_lcore_id])
+                       /* Assigned a new logical core in the loop above. */
+                       qconf = &lcore_queue_conf[rx_lcore_id];
+
+               qconf->rx_port_list[qconf->n_rx_port] = portid;
+               qconf->port_param[qconf->n_rx_port] = &ctrl->vm_ports[portid];
+               qconf->n_rx_port++;
+
+               printf("Lcore %u: RX port %u\n", rx_lcore_id, (unsigned) portid);
+       }
+
+       sigsetup();
+
+       /* indicate that we are ready to forward */
+       ctrl->state = STATE_FWD;
+
+       /* unlock */
+       rte_spinlock_unlock(&ctrl->lock);
+
+       /* launch per-lcore init on every lcore */
+       rte_eal_mp_remote_launch(l2fwd_ivshmem_launch_one_lcore, NULL, CALL_MASTER);
+       RTE_LCORE_FOREACH_SLAVE(lcore_id) {
+               if (rte_eal_wait_lcore(lcore_id) < 0)
+                       return -1;
+       }
+
+       return 0;
+}
diff --git a/examples/l2fwd-ivshmem/host/Makefile b/examples/l2fwd-ivshmem/host/Makefile
new file mode 100644 (file)
index 0000000..a7e7525
--- /dev/null
@@ -0,0 +1,50 @@
+#   BSD LICENSE
+# 
+#   Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
+#   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 Intel Corporation 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.
+
+ifeq ($(RTE_SDK),)
+$(error "Please define RTE_SDK environment variable")
+endif
+
+# Default target, can be overriden by command line or environment
+RTE_TARGET ?= x86_64-ivshmem-linuxapp-gcc
+
+include $(RTE_SDK)/mk/rte.vars.mk
+
+# binary name
+APP = host
+
+# all source are stored in SRCS-y
+SRCS-y := host.c
+
+CFLAGS += -O3
+CFLAGS += $(WERROR_FLAGS)
+
+include $(RTE_SDK)/mk/rte.extapp.mk
diff --git a/examples/l2fwd-ivshmem/host/host.c b/examples/l2fwd-ivshmem/host/host.c
new file mode 100644 (file)
index 0000000..6aaa082
--- /dev/null
@@ -0,0 +1,945 @@
+/*-
+ *   BSD LICENSE
+ * 
+ *   Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
+ *   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 Intel Corporation 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 <unistd.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include <string.h>
+#include <limits.h>
+#include <inttypes.h>
+#include <getopt.h>
+#include <signal.h>
+
+#include <rte_eal.h>
+#include <rte_config.h>
+#include <rte_cycles.h>
+#include <rte_eal_memconfig.h>
+#include <rte_debug.h>
+#include <rte_ether.h>
+#include <rte_ethdev.h>
+#include <rte_string_fns.h>
+#include <rte_ivshmem.h>
+#include <rte_ring.h>
+#include <rte_mempool.h>
+#include <rte_mbuf.h>
+
+#include "../include/common.h"
+
+/*
+ * Configurable number of RX/TX ring descriptors
+ */
+#define RTE_TEST_RX_DESC_DEFAULT 128
+#define RTE_TEST_TX_DESC_DEFAULT 512
+static uint16_t nb_rxd = RTE_TEST_RX_DESC_DEFAULT;
+static uint16_t nb_txd = RTE_TEST_TX_DESC_DEFAULT;
+
+/*
+ * RX and TX Prefetch, Host, and Write-back threshold values should be
+ * carefully set for optimal performance. Consult the network
+ * controller's datasheet and supporting DPDK documentation for guidance
+ * on how these parameters should be set.
+ */
+#define RX_PTHRESH 8 /**< Default values of RX prefetch threshold reg. */
+#define RX_HTHRESH 8 /**< Default values of RX host threshold reg. */
+#define RX_WTHRESH 4 /**< Default values of RX write-back threshold reg. */
+
+/*
+ * These default values are optimized for use with the Intel(R) 82599 10 GbE
+ * Controller and the DPDK ixgbe PMD. Consider using other values for other
+ * network controllers and/or network drivers.
+ */
+#define TX_PTHRESH 36 /**< Default values of TX prefetch threshold reg. */
+#define TX_HTHRESH 0  /**< Default values of TX host threshold reg. */
+#define TX_WTHRESH 0  /**< Default values of TX write-back threshold reg. */
+
+#define BURST_TX_DRAIN_US 100 /* TX drain every ~100us */
+
+/* mask of enabled ports */
+static uint32_t l2fwd_ivshmem_enabled_port_mask = 0;
+
+static struct ether_addr l2fwd_ivshmem_ports_eth_addr[RTE_MAX_ETHPORTS];
+
+#define NB_MBUF   8192
+#define MBUF_SIZE (2048 + sizeof(struct rte_mbuf) + RTE_PKTMBUF_HEADROOM)
+
+#define MAX_RX_QUEUE_PER_LCORE 16
+#define MAX_TX_QUEUE_PER_PORT 16
+struct lcore_queue_conf {
+       unsigned n_rx_port;
+       unsigned rx_port_list[MAX_RX_QUEUE_PER_LCORE];
+       struct vm_port_param * port_param[MAX_RX_QUEUE_PER_LCORE];
+       struct mbuf_table tx_mbufs[RTE_MAX_ETHPORTS];
+       struct mbuf_table rx_mbufs[RTE_MAX_ETHPORTS];
+} __rte_cache_aligned;
+static struct lcore_queue_conf lcore_queue_conf[RTE_MAX_LCORE];
+
+static const struct rte_eth_conf port_conf = {
+       .rxmode = {
+               .split_hdr_size = 0,
+               .header_split   = 0, /**< Header Split disabled */
+               .hw_ip_checksum = 0, /**< IP checksum offload disabled */
+               .hw_vlan_filter = 0, /**< VLAN filtering disabled */
+               .jumbo_frame    = 0, /**< Jumbo Frame Support disabled */
+               .hw_strip_crc   = 0, /**< CRC stripped by hardware */
+       },
+       .txmode = {
+               .mq_mode = ETH_MQ_TX_NONE,
+       },
+};
+
+static const struct rte_eth_rxconf rx_conf = {
+       .rx_thresh = {
+               .pthresh = RX_PTHRESH,
+               .hthresh = RX_HTHRESH,
+               .wthresh = RX_WTHRESH,
+       },
+};
+
+static const struct rte_eth_txconf tx_conf = {
+       .tx_thresh = {
+               .pthresh = TX_PTHRESH,
+               .hthresh = TX_HTHRESH,
+               .wthresh = TX_WTHRESH,
+       },
+       .tx_free_thresh = 0, /* Use PMD default values */
+       .tx_rs_thresh = 0, /* Use PMD default values */
+};
+
+#define METADATA_NAME "l2fwd_ivshmem"
+#define CMDLINE_OPT_FWD_CONF "fwd-conf"
+
+#define QEMU_CMD_FMT "/tmp/ivshmem_qemu_cmdline_%s"
+
+struct port_statistics port_statistics[RTE_MAX_ETHPORTS];
+
+struct rte_mempool * l2fwd_ivshmem_pktmbuf_pool = NULL;
+
+/* Print out statistics on packets dropped */
+static void
+print_stats(void)
+{
+       uint64_t total_packets_dropped, total_packets_tx, total_packets_rx;
+       uint64_t total_vm_packets_dropped, total_vm_packets_tx, total_vm_packets_rx;
+       unsigned portid;
+
+       total_packets_dropped = 0;
+       total_packets_tx = 0;
+       total_packets_rx = 0;
+       total_vm_packets_tx = 0;
+       total_vm_packets_rx = 0;
+
+       const char clr[] = { 27, '[', '2', 'J', '\0' };
+       const char topLeft[] = { 27, '[', '1', ';', '1', 'H','\0' };
+
+               /* Clear screen and move to top left */
+       printf("%s%s", clr, topLeft);
+
+       printf("\nPort statistics ====================================");
+
+       for (portid = 0; portid < RTE_MAX_ETHPORTS; portid++) {
+               /* skip disabled ports */
+               if ((l2fwd_ivshmem_enabled_port_mask & (1 << portid)) == 0)
+                       continue;
+               printf("\nStatistics for port %u ------------------------------"
+                          "\nPackets sent: %24"PRIu64
+                          "\nPackets received: %20"PRIu64
+                          "\nPackets dropped: %21"PRIu64,
+                          portid,
+                          port_statistics[portid].tx,
+                          port_statistics[portid].rx,
+                          port_statistics[portid].dropped);
+
+               total_packets_dropped += port_statistics[portid].dropped;
+               total_packets_tx += port_statistics[portid].tx;
+               total_packets_rx += port_statistics[portid].rx;
+       }
+
+       printf("\nVM statistics ======================================");
+       for (portid = 0; portid < ctrl->nb_ports; portid++) {
+               printf("\nStatistics for port %u ------------------------------"
+                          "\nPackets sent: %24"PRIu64
+                          "\nPackets received: %20"PRIu64,
+                          portid,
+                          ctrl->vm_ports[portid].stats.tx,
+                          ctrl->vm_ports[portid].stats.rx);
+
+               total_vm_packets_dropped += ctrl->vm_ports[portid].stats.dropped;
+               total_vm_packets_tx += ctrl->vm_ports[portid].stats.tx;
+               total_vm_packets_rx += ctrl->vm_ports[portid].stats.rx;
+       }
+       printf("\nAggregate statistics ==============================="
+                          "\nTotal packets sent: %18"PRIu64
+                          "\nTotal packets received: %14"PRIu64
+                          "\nTotal packets dropped: %15"PRIu64
+                          "\nTotal VM packets sent: %15"PRIu64
+                          "\nTotal VM packets received: %11"PRIu64,
+                          total_packets_tx,
+                          total_packets_rx,
+                          total_packets_dropped,
+                          total_vm_packets_tx,
+                          total_vm_packets_rx);
+       printf("\n====================================================\n");
+}
+
+static int
+print_to_file(const char *cmdline, const char *config_name)
+{
+       FILE *file;
+       char path[PATH_MAX];
+
+       rte_snprintf(path, sizeof(path), QEMU_CMD_FMT, config_name);
+       file = fopen(path, "w");
+       if (file == NULL) {
+               RTE_LOG(ERR, L2FWD_IVSHMEM, "Could not open '%s' \n", path);
+               return -1;
+       }
+
+       RTE_LOG(DEBUG, L2FWD_IVSHMEM, "QEMU command line for config '%s': %s \n",
+                       config_name, cmdline);
+
+       fprintf(file, "%s\n", cmdline);
+       fclose(file);
+       return 0;
+}
+
+static int
+generate_ivshmem_cmdline(const char *config_name)
+{
+       char cmdline[PATH_MAX];
+       if (rte_ivshmem_metadata_cmdline_generate(cmdline, sizeof(cmdline),
+                       config_name) < 0)
+               return -1;
+
+       if (print_to_file(cmdline, config_name) < 0)
+               return -1;
+
+       rte_ivshmem_metadata_dump(config_name);
+       return 0;
+}
+
+/* display usage */
+static void
+l2fwd_ivshmem_usage(const char *prgname)
+{
+       printf("%s [EAL options] -- -p PORTMASK [-q NQ -T PERIOD]\n"
+                  "  -p PORTMASK: hexadecimal bitmask of ports to configure\n"
+                  "  -q NQ: number of queue (=ports) per lcore (default is 1)\n"
+                  "  -T PERIOD: statistics will be refreshed each PERIOD seconds "
+                      "(0 to disable, 10 default, 86400 maximum)\n",
+              prgname);
+}
+
+static unsigned int
+l2fwd_ivshmem_parse_nqueue(const char *q_arg)
+{
+       char *end = NULL;
+       unsigned long n;
+
+       /* parse hexadecimal string */
+       n = strtoul(q_arg, &end, 10);
+       if ((q_arg[0] == '\0') || (end == NULL) || (*end != '\0'))
+               return 0;
+       if (n == 0)
+               return 0;
+       if (n >= MAX_RX_QUEUE_PER_LCORE)
+               return 0;
+
+       return n;
+}
+
+static int
+l2fwd_ivshmem_parse_portmask(const char *portmask)
+{
+       char *end = NULL;
+       unsigned long pm;
+
+       /* parse hexadecimal string */
+       pm = strtoul(portmask, &end, 16);
+       if ((portmask[0] == '\0') || (end == NULL) || (*end != '\0'))
+               return -1;
+
+       if (pm == 0)
+               return -1;
+
+       return pm;
+}
+
+static int
+l2fwd_ivshmem_parse_timer_period(const char *q_arg)
+{
+       char *end = NULL;
+       int n;
+
+       /* parse number string */
+       n = strtol(q_arg, &end, 10);
+       if ((q_arg[0] == '\0') || (end == NULL) || (*end != '\0'))
+               return -1;
+       if (n >= MAX_TIMER_PERIOD)
+               return -1;
+
+       return n;
+}
+
+/* Parse the argument given in the command line of the application */
+static int
+l2fwd_ivshmem_parse_args(int argc, char **argv)
+{
+       int opt, ret;
+       char **argvopt;
+       int option_index;
+       char *prgname = argv[0];
+       static struct option lgopts[] = {
+                       {CMDLINE_OPT_FWD_CONF, 1, 0, 0},
+               {NULL, 0, 0, 0}
+       };
+
+       argvopt = argv;
+
+       while ((opt = getopt_long(argc, argvopt, "q:p:T:",
+                                 lgopts, &option_index)) != EOF) {
+
+               switch (opt) {
+               /* portmask */
+               case 'p':
+                       l2fwd_ivshmem_enabled_port_mask = l2fwd_ivshmem_parse_portmask(optarg);
+                       if (l2fwd_ivshmem_enabled_port_mask == 0) {
+                               printf("invalid portmask\n");
+                               l2fwd_ivshmem_usage(prgname);
+                               return -1;
+                       }
+                       break;
+
+               /* nqueue */
+               case 'q':
+                       l2fwd_ivshmem_rx_queue_per_lcore = l2fwd_ivshmem_parse_nqueue(optarg);
+                       if (l2fwd_ivshmem_rx_queue_per_lcore == 0) {
+                               printf("invalid queue number\n");
+                               l2fwd_ivshmem_usage(prgname);
+                               return -1;
+                       }
+                       break;
+
+               /* timer period */
+               case 'T':
+                       timer_period = l2fwd_ivshmem_parse_timer_period(optarg) * 1000 * TIMER_MILLISECOND;
+                       if (timer_period < 0) {
+                               printf("invalid timer period\n");
+                               l2fwd_ivshmem_usage(prgname);
+                               return -1;
+                       }
+                       break;
+
+               /* long options */
+               case 0:
+                       l2fwd_ivshmem_usage(prgname);
+                       return -1;
+
+               default:
+                       l2fwd_ivshmem_usage(prgname);
+                       return -1;
+               }
+       }
+
+       if (optind >= 0)
+               argv[optind-1] = prgname;
+
+       ret = optind-1;
+       optind = 0; /* reset getopt lib */
+       return ret;
+}
+
+/* Check the link status of all ports in up to 9s, and print them finally */
+static void
+check_all_ports_link_status(uint8_t port_num, uint32_t port_mask)
+{
+#define CHECK_INTERVAL 100 /* 100ms */
+#define MAX_CHECK_TIME 90 /* 9s (90 * 100ms) in total */
+       uint8_t portid, count, all_ports_up, print_flag = 0;
+       struct rte_eth_link link;
+
+       printf("\nChecking link status");
+       fflush(stdout);
+       for (count = 0; count <= MAX_CHECK_TIME; count++) {
+               all_ports_up = 1;
+               for (portid = 0; portid < port_num; portid++) {
+                       if ((port_mask & (1 << portid)) == 0)
+                               continue;
+                       memset(&link, 0, sizeof(link));
+                       rte_eth_link_get_nowait(portid, &link);
+                       /* print link status if flag set */
+                       if (print_flag == 1) {
+                               if (link.link_status)
+                                       printf("Port %d Link Up - speed %u "
+                                               "Mbps - %s\n", (uint8_t)portid,
+                                               (unsigned)link.link_speed,
+                               (link.link_duplex == ETH_LINK_FULL_DUPLEX) ?
+                                       ("full-duplex") : ("half-duplex\n"));
+                               else
+                                       printf("Port %d Link Down\n",
+                                               (uint8_t)portid);
+                               continue;
+                       }
+                       /* clear all_ports_up flag if any link down */
+                       if (link.link_status == 0) {
+                               all_ports_up = 0;
+                               break;
+                       }
+               }
+               /* after finally printing all link status, get out */
+               if (print_flag == 1)
+                       break;
+
+               if (all_ports_up == 0) {
+                       printf(".");
+                       fflush(stdout);
+                       rte_delay_ms(CHECK_INTERVAL);
+               }
+
+               /* set the print_flag if all ports up or timeout */
+               if (all_ports_up == 1 || count == (MAX_CHECK_TIME - 1)) {
+                       print_flag = 1;
+                       printf("done\n");
+               }
+       }
+}
+
+/* Send the burst of packets on an output interface */
+static int
+l2fwd_ivshmem_send_burst(struct lcore_queue_conf *qconf, unsigned n, uint8_t port)
+{
+       struct rte_mbuf **m_table;
+       unsigned ret;
+       unsigned queueid =0;
+
+       m_table = (struct rte_mbuf **)qconf->tx_mbufs[port].m_table;
+
+       ret = rte_eth_tx_burst(port, (uint16_t) queueid, m_table, (uint16_t) n);
+       port_statistics[port].tx += ret;
+       if (unlikely(ret < n)) {
+               port_statistics[port].dropped += (n - ret);
+               do {
+                       rte_pktmbuf_free(m_table[ret]);
+               } while (++ret < n);
+       }
+
+       return 0;
+}
+
+/* Enqueue packets for TX and prepare them to be sent on the network */
+static int
+l2fwd_ivshmem_send_packet(struct rte_mbuf *m, uint8_t port)
+{
+       unsigned lcore_id, len;
+       struct lcore_queue_conf *qconf;
+
+       lcore_id = rte_lcore_id();
+
+       qconf = &lcore_queue_conf[lcore_id];
+       len = qconf->tx_mbufs[port].len;
+       qconf->tx_mbufs[port].m_table[len] = m;
+       len++;
+
+       /* enough pkts to be sent */
+       if (unlikely(len == MAX_PKT_BURST)) {
+               l2fwd_ivshmem_send_burst(qconf, MAX_PKT_BURST, port);
+               len = 0;
+       }
+
+       qconf->tx_mbufs[port].len = len;
+       return 0;
+}
+
+static int
+l2fwd_ivshmem_receive_burst(struct lcore_queue_conf *qconf, unsigned portid,
+               unsigned vm_port)
+{
+       struct rte_mbuf ** m;
+       struct rte_ring * rx;
+       unsigned len, pkt_idx;
+
+       m = qconf->rx_mbufs[portid].m_table;
+       len = qconf->rx_mbufs[portid].len;
+       rx = qconf->port_param[vm_port]->rx_ring;
+
+       /* if enqueueing failed, ring is probably full, so drop the packets */
+       if (rte_ring_enqueue_bulk(rx, (void**) m, len) < 0) {
+               port_statistics[portid].dropped += len;
+
+               pkt_idx = 0;
+               do {
+                       rte_pktmbuf_free(m[pkt_idx]);
+               } while (++pkt_idx < len);
+       }
+       else
+               /* increment rx stats by however many packets we managed to receive */
+               port_statistics[portid].rx += len;
+
+       return 0;
+}
+
+/* Enqueue packets for RX and prepare them to be sent to VM */
+static int
+l2fwd_ivshmem_receive_packets(struct rte_mbuf ** m, unsigned n, unsigned portid,
+               unsigned vm_port)
+{
+       unsigned lcore_id, len, pkt_idx;
+       struct lcore_queue_conf *qconf;
+
+       lcore_id = rte_lcore_id();
+
+       qconf = &lcore_queue_conf[lcore_id];
+
+       len = qconf->rx_mbufs[portid].len;
+       pkt_idx = 0;
+
+       /* enqueue packets */
+       while (pkt_idx < n && len < MAX_PKT_BURST * 2) {
+               qconf->rx_mbufs[portid].m_table[len++] = m[pkt_idx++];
+       }
+
+       /* increment queue len by however many packets we managed to receive */
+       qconf->rx_mbufs[portid].len += pkt_idx;
+
+       /* drop the unreceived packets */
+       if (unlikely(pkt_idx < n)) {
+               port_statistics[portid].dropped += n - pkt_idx;
+               do {
+                       rte_pktmbuf_free(m[pkt_idx]);
+               } while (++pkt_idx < n);
+       }
+
+       /* drain the queue halfway through the maximum capacity */
+       if (unlikely(qconf->rx_mbufs[portid].len >= MAX_PKT_BURST))
+               l2fwd_ivshmem_receive_burst(qconf, portid, vm_port);
+
+       return 0;
+}
+
+/* loop for host forwarding mode.
+ * the data flow is as follows:
+ *  1) get packets from TX queue and send it out from a given port
+ *  2) RX packets from given port and enqueue them on RX ring
+ *  3) dequeue packets from TX ring and put them on TX queue for a given port
+ */
+static void
+fwd_loop(void)
+{
+       struct rte_mbuf *pkts_burst[MAX_PKT_BURST * 2];
+       struct rte_mbuf *m;
+       unsigned lcore_id;
+       uint64_t prev_tsc, diff_tsc, cur_tsc, timer_tsc;
+       unsigned i, j, portid, nb_rx;
+       struct lcore_queue_conf *qconf;
+       struct rte_ring *tx;
+       const uint64_t drain_tsc = (rte_get_tsc_hz() + US_PER_S - 1) / US_PER_S * BURST_TX_DRAIN_US;
+
+       prev_tsc = 0;
+       timer_tsc = 0;
+
+       lcore_id = rte_lcore_id();
+       qconf = &lcore_queue_conf[lcore_id];
+
+       if (qconf->n_rx_port == 0) {
+               RTE_LOG(INFO, L2FWD_IVSHMEM, "lcore %u has nothing to do\n", lcore_id);
+               return;
+       }
+
+       RTE_LOG(INFO, L2FWD_IVSHMEM, "entering main loop on lcore %u\n", lcore_id);
+
+       for (i = 0; i < qconf->n_rx_port; i++) {
+
+               portid = qconf->rx_port_list[i];
+               RTE_LOG(INFO, L2FWD_IVSHMEM, " -- lcoreid=%u portid=%u\n", lcore_id,
+                       portid);
+       }
+
+       while (ctrl->state == STATE_FWD) {
+
+               cur_tsc = rte_rdtsc();
+
+               /*
+                * Burst queue drain
+                */
+               diff_tsc = cur_tsc - prev_tsc;
+               if (unlikely(diff_tsc > drain_tsc)) {
+
+                       /*
+                        * TX
+                        */
+                       for (portid = 0; portid < RTE_MAX_ETHPORTS; portid++) {
+                               if (qconf->tx_mbufs[portid].len == 0)
+                                       continue;
+                               l2fwd_ivshmem_send_burst(qconf,
+                                                qconf->tx_mbufs[portid].len,
+                                                (uint8_t) portid);
+                               qconf->tx_mbufs[portid].len = 0;
+                       }
+
+                       /*
+                        * RX
+                        */
+                       for (i = 0; i < qconf->n_rx_port; i++) {
+                               portid = qconf->rx_port_list[i];
+                               if (qconf->rx_mbufs[portid].len == 0)
+                                       continue;
+                               l2fwd_ivshmem_receive_burst(qconf, portid, i);
+                               qconf->rx_mbufs[portid].len = 0;
+                       }
+
+                       /* if timer is enabled */
+                       if (timer_period > 0) {
+
+                               /* advance the timer */
+                               timer_tsc += diff_tsc;
+
+                               /* if timer has reached its timeout */
+                               if (unlikely(timer_tsc >= (uint64_t) timer_period)) {
+
+                                       /* do this only on master core */
+                                       if (lcore_id == rte_get_master_lcore()) {
+                                               print_stats();
+                                               /* reset the timer */
+                                               timer_tsc = 0;
+                                       }
+                               }
+                       }
+
+                       prev_tsc = cur_tsc;
+               }
+
+               /*
+                * packet RX and forwarding
+                */
+               for (i = 0; i < qconf->n_rx_port; i++) {
+
+                       /* RX packets from port and put them on RX ring */
+                       portid = qconf->rx_port_list[i];
+                       nb_rx = rte_eth_rx_burst((uint8_t) portid, 0,
+                                                pkts_burst, MAX_PKT_BURST);
+
+                       if (nb_rx != 0)
+                               l2fwd_ivshmem_receive_packets(pkts_burst, nb_rx, portid, i);
+
+                       /* dequeue packets from TX ring and send them to TX queue */
+                       tx = qconf->port_param[i]->tx_ring;
+
+                       nb_rx = rte_ring_count(tx);
+
+                       nb_rx = RTE_MIN(nb_rx, (unsigned) MAX_PKT_BURST);
+
+                       if (nb_rx == 0)
+                               continue;
+
+                       /* should not happen */
+                       if (unlikely(rte_ring_dequeue_bulk(tx, (void**) pkts_burst, nb_rx) < 0)) {
+                               ctrl->state = STATE_FAIL;
+                               return;
+                       }
+
+                       for (j = 0; j < nb_rx; j++) {
+                               m = pkts_burst[j];
+                               l2fwd_ivshmem_send_packet(m, portid);
+                       }
+               }
+       }
+}
+
+static int
+l2fwd_ivshmem_launch_one_lcore(__attribute__((unused)) void *dummy)
+{
+       fwd_loop();
+       return 0;
+}
+
+int main(int argc, char **argv)
+{
+       char name[RTE_RING_NAMESIZE];
+       struct rte_ring *r;
+       struct lcore_queue_conf *qconf;
+       struct rte_eth_dev_info dev_info;
+       uint8_t portid, port_nr;
+       uint8_t nb_ports, nb_ports_available;
+       uint8_t nb_ports_in_mask;
+       int ret;
+       unsigned lcore_id, rx_lcore_id;
+
+       /* init EAL */
+       ret = rte_eal_init(argc, argv);
+       if (ret < 0)
+               rte_exit(EXIT_FAILURE, "Invalid EAL arguments\n");
+       argc -= ret;
+       argv += ret;
+
+       /* parse application arguments (after the EAL ones) */
+       ret = l2fwd_ivshmem_parse_args(argc, argv);
+       if (ret < 0)
+               rte_exit(EXIT_FAILURE, "Invalid l2fwd-ivshmem arguments\n");
+
+       /* create a shared mbuf pool */
+       l2fwd_ivshmem_pktmbuf_pool =
+               rte_mempool_create(MBUF_MP_NAME, NB_MBUF,
+                                  MBUF_SIZE, 32,
+                                  sizeof(struct rte_pktmbuf_pool_private),
+                                  rte_pktmbuf_pool_init, NULL,
+                                  rte_pktmbuf_init, NULL,
+                                  rte_socket_id(), 0);
+       if (l2fwd_ivshmem_pktmbuf_pool == NULL)
+               rte_exit(EXIT_FAILURE, "Cannot init mbuf pool\n");
+
+       /* init driver(s) */
+       if (rte_pmd_init_all() < 0)
+               rte_exit(EXIT_FAILURE, "Cannot init pmd\n");
+
+       if (rte_eal_pci_probe() < 0)
+               rte_exit(EXIT_FAILURE, "Cannot probe PCI\n");
+
+       nb_ports = rte_eth_dev_count();
+       if (nb_ports == 0)
+               rte_exit(EXIT_FAILURE, "No Ethernet ports - bye\n");
+
+       if (nb_ports > RTE_MAX_ETHPORTS)
+               nb_ports = RTE_MAX_ETHPORTS;
+
+       /*
+        * reserve memzone to communicate with VMs - we cannot use rte_malloc here
+        * because while it is technically possible, it is a very bad idea to share
+        * the heap between two primary processes.
+        */
+       ctrl_mz = rte_memzone_reserve(CTRL_MZ_NAME, sizeof(struct ivshmem_ctrl),
+                       SOCKET_ID_ANY, 0);
+       if (ctrl_mz == NULL)
+               rte_exit(EXIT_FAILURE, "Cannot reserve control memzone\n");
+       ctrl = (struct ivshmem_ctrl*) ctrl_mz->addr;
+
+       memset(ctrl, 0, sizeof(struct ivshmem_ctrl));
+
+       /*
+        * Each port is assigned an output port.
+        */
+       nb_ports_in_mask = 0;
+       for (portid = 0; portid < nb_ports; portid++) {
+               /* skip ports that are not enabled */
+               if ((l2fwd_ivshmem_enabled_port_mask & (1 << portid)) == 0)
+                       continue;
+               if (portid % 2) {
+                       ctrl->vm_ports[nb_ports_in_mask].dst = &ctrl->vm_ports[nb_ports_in_mask-1];
+                       ctrl->vm_ports[nb_ports_in_mask-1].dst = &ctrl->vm_ports[nb_ports_in_mask];
+               }
+
+               nb_ports_in_mask++;
+
+               rte_eth_dev_info_get(portid, &dev_info);
+       }
+       if (nb_ports_in_mask % 2) {
+               printf("Notice: odd number of ports in portmask.\n");
+               ctrl->vm_ports[nb_ports_in_mask-1].dst =
+                               &ctrl->vm_ports[nb_ports_in_mask-1];
+       }
+
+       rx_lcore_id = 0;
+       qconf = NULL;
+
+       printf("Initializing ports configuration...\n");
+
+       nb_ports_available = nb_ports;
+
+       /* Initialise each port */
+       for (portid = 0; portid < nb_ports; portid++) {
+
+               /* skip ports that are not enabled */
+               if ((l2fwd_ivshmem_enabled_port_mask & (1 << portid)) == 0) {
+                       printf("Skipping disabled port %u\n", (unsigned) portid);
+                       nb_ports_available--;
+                       continue;
+               }
+
+               /* init port */
+               printf("Initializing port %u... ", (unsigned) portid);
+               fflush(stdout);
+               ret = rte_eth_dev_configure(portid, 1, 1, &port_conf);
+               if (ret < 0)
+                       rte_exit(EXIT_FAILURE, "Cannot configure device: err=%d, port=%u\n",
+                                 ret, (unsigned) portid);
+
+               rte_eth_macaddr_get(portid,&l2fwd_ivshmem_ports_eth_addr[portid]);
+
+               /* init one RX queue */
+               fflush(stdout);
+               ret = rte_eth_rx_queue_setup(portid, 0, nb_rxd,
+                                                rte_eth_dev_socket_id(portid), &rx_conf,
+                                                l2fwd_ivshmem_pktmbuf_pool);
+               if (ret < 0)
+                       rte_exit(EXIT_FAILURE, "rte_eth_rx_queue_setup:err=%d, port=%u\n",
+                                 ret, (unsigned) portid);
+
+               /* init one TX queue on each port */
+               fflush(stdout);
+               ret = rte_eth_tx_queue_setup(portid, 0, nb_txd,
+                               rte_eth_dev_socket_id(portid), &tx_conf);
+               if (ret < 0)
+                       rte_exit(EXIT_FAILURE, "rte_eth_tx_queue_setup:err=%d, port=%u\n",
+                               ret, (unsigned) portid);
+
+               /* Start device */
+               ret = rte_eth_dev_start(portid);
+               if (ret < 0)
+                       rte_exit(EXIT_FAILURE, "rte_eth_dev_start:err=%d, port=%u\n",
+                                 ret, (unsigned) portid);
+
+               printf("done: \n");
+
+               rte_eth_promiscuous_enable(portid);
+
+               printf("Port %u, MAC address: %02X:%02X:%02X:%02X:%02X:%02X\n\n",
+                               (unsigned) portid,
+                               l2fwd_ivshmem_ports_eth_addr[portid].addr_bytes[0],
+                               l2fwd_ivshmem_ports_eth_addr[portid].addr_bytes[1],
+                               l2fwd_ivshmem_ports_eth_addr[portid].addr_bytes[2],
+                               l2fwd_ivshmem_ports_eth_addr[portid].addr_bytes[3],
+                               l2fwd_ivshmem_ports_eth_addr[portid].addr_bytes[4],
+                               l2fwd_ivshmem_ports_eth_addr[portid].addr_bytes[5]);
+
+               /* initialize port stats */
+               memset(&port_statistics, 0, sizeof(port_statistics));
+       }
+
+       if (!nb_ports_available) {
+               rte_exit(EXIT_FAILURE,
+                       "All available ports are disabled. Please set portmask.\n");
+       }
+       port_nr = 0;
+
+       /* Initialize the port/queue configuration of each logical core */
+       for (portid = 0; portid < nb_ports; portid++) {
+               if ((l2fwd_ivshmem_enabled_port_mask & (1 << portid)) == 0)
+                       continue;
+
+               /* get the lcore_id for this port */
+               while (rte_lcore_is_enabled(rx_lcore_id) == 0 ||
+                          lcore_queue_conf[rx_lcore_id].n_rx_port ==
+                                          l2fwd_ivshmem_rx_queue_per_lcore) {
+                       rx_lcore_id++;
+                       if (rx_lcore_id >= RTE_MAX_LCORE)
+                               rte_exit(EXIT_FAILURE, "Not enough cores\n");
+               }
+
+               if (qconf != &lcore_queue_conf[rx_lcore_id])
+                       /* Assigned a new logical core in the loop above. */
+                       qconf = &lcore_queue_conf[rx_lcore_id];
+
+
+               rte_eth_macaddr_get(portid, &ctrl->vm_ports[port_nr].ethaddr);
+
+               qconf->rx_port_list[qconf->n_rx_port] = portid;
+               qconf->port_param[qconf->n_rx_port] = &ctrl->vm_ports[port_nr];
+               qconf->n_rx_port++;
+               port_nr++;
+               printf("Lcore %u: RX port %u\n", rx_lcore_id, (unsigned) portid);
+       }
+
+       check_all_ports_link_status(nb_ports_available, l2fwd_ivshmem_enabled_port_mask);
+
+       /* create rings for each VM port (several ports can be on the same VM).
+        * note that we store the pointers in ctrl - that way, they are the same
+        * and valid across all VMs because ctrl is also in DPDK memory */
+       for (portid = 0; portid < nb_ports_available; portid++) {
+
+               /* RX ring. SP/SC because it's only used by host and a single VM */
+               rte_snprintf(name, sizeof(name), "%s%i", RX_RING_PREFIX, portid);
+               r = rte_ring_create(name, NB_MBUF,
+                               SOCKET_ID_ANY, RING_F_SP_ENQ | RING_F_SC_DEQ);
+               if (r == NULL)
+                       rte_exit(EXIT_FAILURE, "Cannot create ring %s\n", name);
+
+               ctrl->vm_ports[portid].rx_ring = r;
+
+               /* TX ring. SP/SC because it's only used by host and a single VM */
+               rte_snprintf(name, sizeof(name), "%s%i", TX_RING_PREFIX, portid);
+               r = rte_ring_create(name, NB_MBUF,
+                               SOCKET_ID_ANY, RING_F_SP_ENQ | RING_F_SC_DEQ);
+               if (r == NULL)
+                       rte_exit(EXIT_FAILURE, "Cannot create ring %s\n", name);
+
+               ctrl->vm_ports[portid].tx_ring = r;
+       }
+
+       /* create metadata, output cmdline */
+       if (rte_ivshmem_metadata_create(METADATA_NAME) < 0)
+               rte_exit(EXIT_FAILURE, "Cannot create IVSHMEM metadata\n");
+
+       if (rte_ivshmem_metadata_add_memzone(ctrl_mz, METADATA_NAME))
+               rte_exit(EXIT_FAILURE, "Cannot add memzone to IVSHMEM metadata\n");
+
+       if (rte_ivshmem_metadata_add_mempool(l2fwd_ivshmem_pktmbuf_pool, METADATA_NAME))
+               rte_exit(EXIT_FAILURE, "Cannot add mbuf mempool to IVSHMEM metadata\n");
+
+       for (portid = 0; portid < nb_ports_available; portid++) {
+               if (rte_ivshmem_metadata_add_ring(ctrl->vm_ports[portid].rx_ring,
+                               METADATA_NAME) < 0)
+                       rte_exit(EXIT_FAILURE, "Cannot add ring %s to IVSHMEM metadata\n",
+                                       ctrl->vm_ports[portid].rx_ring->name);
+               if (rte_ivshmem_metadata_add_ring(ctrl->vm_ports[portid].tx_ring,
+                               METADATA_NAME) < 0)
+                       rte_exit(EXIT_FAILURE, "Cannot add ring %s to IVSHMEM metadata\n",
+                                       ctrl->vm_ports[portid].tx_ring->name);
+       }
+       generate_ivshmem_cmdline(METADATA_NAME);
+
+       ctrl->nb_ports = nb_ports_available;
+
+       printf("Waiting for VM to initialize...\n");
+
+       /* wait for VM to initialize */
+       while (ctrl->state != STATE_FWD) {
+               if (ctrl->state == STATE_FAIL)
+                       rte_exit(EXIT_FAILURE, "VM reported failure\n");
+
+               sleep(1);
+       }
+
+       printf("Done!\n");
+
+       sigsetup();
+
+       /* launch per-lcore init on every lcore */
+       rte_eal_mp_remote_launch(l2fwd_ivshmem_launch_one_lcore, NULL, CALL_MASTER);
+       RTE_LCORE_FOREACH_SLAVE(lcore_id) {
+               if (rte_eal_wait_lcore(lcore_id) < 0)
+                       return -1;
+       }
+
+       if (ctrl->state == STATE_FAIL)
+               rte_exit(EXIT_FAILURE, "VM reported failure\n");
+
+       return 0;
+}
diff --git a/examples/l2fwd-ivshmem/include/common.h b/examples/l2fwd-ivshmem/include/common.h
new file mode 100644 (file)
index 0000000..49ac26d
--- /dev/null
@@ -0,0 +1,111 @@
+/*-
+ *   BSD LICENSE
+ * 
+ *   Copyright(c) 2010-2014 Intel Corporation. All rights reserved.
+ *   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 Intel Corporation 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 _IVSHMEM_COMMON_H_
+#define _IVSHMEM_COMMON_H_
+
+#define RTE_LOGTYPE_L2FWD_IVSHMEM RTE_LOGTYPE_USER1
+
+#define CTRL_MZ_NAME "CTRL_MEMZONE"
+#define MBUF_MP_NAME "MBUF_MEMPOOL"
+#define RX_RING_PREFIX "RX_"
+#define TX_RING_PREFIX "TX_"
+
+/* A tsc-based timer responsible for triggering statistics printout */
+#define TIMER_MILLISECOND 2000000ULL /* around 1ms at 2 Ghz */
+#define MAX_TIMER_PERIOD 86400 /* 1 day max */
+static int64_t timer_period = 10 * TIMER_MILLISECOND * 1000; /* default period is 10 seconds */
+
+#define DIM(x)\
+       (sizeof(x)/sizeof(x)[0])
+
+#define MAX_PKT_BURST 32
+
+const struct rte_memzone * ctrl_mz;
+
+enum l2fwd_state {
+       STATE_NONE = 0,
+       STATE_FWD,
+       STATE_EXIT,
+       STATE_FAIL
+};
+
+/* Per-port statistics struct */
+struct port_statistics {
+       uint64_t tx;
+       uint64_t rx;
+       uint64_t dropped;
+} __rte_cache_aligned;
+
+struct mbuf_table {
+       unsigned len;
+       struct rte_mbuf *m_table[MAX_PKT_BURST * 2]; /**< allow up to two bursts */
+};
+
+struct vm_port_param {
+       struct rte_ring * rx_ring;         /**< receiving ring for current port */
+       struct rte_ring * tx_ring;         /**< transmitting ring for current port */
+       struct vm_port_param * dst;        /**< current port's destination port */
+       volatile struct port_statistics stats;      /**< statistics for current port */
+       struct ether_addr ethaddr;         /**< Ethernet address of the port */
+};
+
+/* control structure, to synchronize host and VM */
+struct ivshmem_ctrl {
+       rte_spinlock_t lock;
+       uint8_t nb_ports;                /**< total nr of ports */
+       volatile enum l2fwd_state state; /**< report state */
+       struct vm_port_param vm_ports[RTE_MAX_ETHPORTS];
+};
+
+struct ivshmem_ctrl * ctrl;
+
+static unsigned int l2fwd_ivshmem_rx_queue_per_lcore = 1;
+
+static void sighandler(int __rte_unused s)
+{
+       ctrl->state = STATE_EXIT;
+}
+
+static void sigsetup(void)
+{
+          struct sigaction sigIntHandler;
+
+          sigIntHandler.sa_handler = sighandler;
+          sigemptyset(&sigIntHandler.sa_mask);
+          sigIntHandler.sa_flags = 0;
+
+          sigaction(SIGINT, &sigIntHandler, NULL);
+}
+
+#endif /* _IVSHMEM_COMMON_H_ */