summaryrefslogtreecommitdiffstats
path: root/src/seastar/dpdk/test/test/test_kni.c
diff options
context:
space:
mode:
Diffstat (limited to 'src/seastar/dpdk/test/test/test_kni.c')
-rw-r--r--src/seastar/dpdk/test/test/test_kni.c636
1 files changed, 636 insertions, 0 deletions
diff --git a/src/seastar/dpdk/test/test/test_kni.c b/src/seastar/dpdk/test/test/test_kni.c
new file mode 100644
index 00000000..db17fdf3
--- /dev/null
+++ b/src/seastar/dpdk/test/test/test_kni.c
@@ -0,0 +1,636 @@
+/*-
+ * 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 <stdint.h>
+#include <unistd.h>
+#include <string.h>
+#include <sys/wait.h>
+
+#include "test.h"
+
+#include <rte_string_fns.h>
+#include <rte_mempool.h>
+#include <rte_ethdev.h>
+#include <rte_cycles.h>
+#include <rte_kni.h>
+
+#define NB_MBUF 8192
+#define MAX_PACKET_SZ 2048
+#define MBUF_DATA_SZ (MAX_PACKET_SZ + RTE_PKTMBUF_HEADROOM)
+#define PKT_BURST_SZ 32
+#define MEMPOOL_CACHE_SZ PKT_BURST_SZ
+#define SOCKET 0
+#define NB_RXD 128
+#define NB_TXD 512
+#define KNI_TIMEOUT_MS 5000 /* ms */
+
+#define IFCONFIG "/sbin/ifconfig "
+#define TEST_KNI_PORT "test_kni_port"
+#define KNI_TEST_MAX_PORTS 4
+/* The threshold number of mbufs to be transmitted or received. */
+#define KNI_NUM_MBUF_THRESHOLD 100
+static int kni_pkt_mtu = 0;
+
+struct test_kni_stats {
+ volatile uint64_t ingress;
+ volatile uint64_t egress;
+};
+
+static const struct rte_eth_rxconf rx_conf = {
+ .rx_thresh = {
+ .pthresh = 8,
+ .hthresh = 8,
+ .wthresh = 4,
+ },
+ .rx_free_thresh = 0,
+};
+
+static const struct rte_eth_txconf tx_conf = {
+ .tx_thresh = {
+ .pthresh = 36,
+ .hthresh = 0,
+ .wthresh = 0,
+ },
+ .tx_free_thresh = 0,
+ .tx_rs_thresh = 0,
+};
+
+static const struct rte_eth_conf port_conf = {
+ .rxmode = {
+ .header_split = 0,
+ .hw_ip_checksum = 0,
+ .hw_vlan_filter = 0,
+ .jumbo_frame = 0,
+ .hw_strip_crc = 1,
+ },
+ .txmode = {
+ .mq_mode = ETH_DCB_NONE,
+ },
+};
+
+static struct rte_kni_ops kni_ops = {
+ .change_mtu = NULL,
+ .config_network_if = NULL,
+};
+
+static unsigned lcore_master, lcore_ingress, lcore_egress;
+static struct rte_kni *test_kni_ctx;
+static struct test_kni_stats stats;
+
+static volatile uint32_t test_kni_processing_flag;
+
+static struct rte_mempool *
+test_kni_create_mempool(void)
+{
+ struct rte_mempool * mp;
+
+ mp = rte_mempool_lookup("kni_mempool");
+ if (!mp)
+ mp = rte_pktmbuf_pool_create("kni_mempool",
+ NB_MBUF,
+ MEMPOOL_CACHE_SZ, 0, MBUF_DATA_SZ,
+ SOCKET);
+
+ return mp;
+}
+
+static struct rte_mempool *
+test_kni_lookup_mempool(void)
+{
+ return rte_mempool_lookup("kni_mempool");
+}
+/* Callback for request of changing MTU */
+static int
+kni_change_mtu(uint8_t port_id, unsigned new_mtu)
+{
+ printf("Change MTU of port %d to %u\n", port_id, new_mtu);
+ kni_pkt_mtu = new_mtu;
+ printf("Change MTU of port %d to %i successfully.\n",
+ port_id, kni_pkt_mtu);
+ return 0;
+}
+/**
+ * This loop fully tests the basic functions of KNI. e.g. transmitting,
+ * receiving to, from kernel space, and kernel requests.
+ *
+ * This is the loop to transmit/receive mbufs to/from kernel interface with
+ * supported by KNI kernel module. The ingress lcore will allocate mbufs and
+ * transmit them to kernel space; while the egress lcore will receive the mbufs
+ * from kernel space and free them.
+ * On the master lcore, several commands will be run to check handling the
+ * kernel requests. And it will finally set the flag to exit the KNI
+ * transmitting/receiving to/from the kernel space.
+ *
+ * Note: To support this testing, the KNI kernel module needs to be insmodded
+ * in one of its loopback modes.
+ */
+static int
+test_kni_loop(__rte_unused void *arg)
+{
+ int ret = 0;
+ unsigned nb_rx, nb_tx, num, i;
+ const unsigned lcore_id = rte_lcore_id();
+ struct rte_mbuf *pkts_burst[PKT_BURST_SZ];
+
+ if (lcore_id == lcore_master) {
+ rte_delay_ms(KNI_TIMEOUT_MS);
+ /* tests of handling kernel request */
+ if (system(IFCONFIG TEST_KNI_PORT" up") == -1)
+ ret = -1;
+ if (system(IFCONFIG TEST_KNI_PORT" mtu 1400") == -1)
+ ret = -1;
+ if (system(IFCONFIG TEST_KNI_PORT" down") == -1)
+ ret = -1;
+ rte_delay_ms(KNI_TIMEOUT_MS);
+ test_kni_processing_flag = 1;
+ } else if (lcore_id == lcore_ingress) {
+ struct rte_mempool *mp = test_kni_lookup_mempool();
+
+ if (mp == NULL)
+ return -1;
+
+ while (1) {
+ if (test_kni_processing_flag)
+ break;
+
+ for (nb_rx = 0; nb_rx < PKT_BURST_SZ; nb_rx++) {
+ pkts_burst[nb_rx] = rte_pktmbuf_alloc(mp);
+ if (!pkts_burst[nb_rx])
+ break;
+ }
+
+ num = rte_kni_tx_burst(test_kni_ctx, pkts_burst,
+ nb_rx);
+ stats.ingress += num;
+ rte_kni_handle_request(test_kni_ctx);
+ if (num < nb_rx) {
+ for (i = num; i < nb_rx; i++) {
+ rte_pktmbuf_free(pkts_burst[i]);
+ }
+ }
+ rte_delay_ms(10);
+ }
+ } else if (lcore_id == lcore_egress) {
+ while (1) {
+ if (test_kni_processing_flag)
+ break;
+ num = rte_kni_rx_burst(test_kni_ctx, pkts_burst,
+ PKT_BURST_SZ);
+ stats.egress += num;
+ for (nb_tx = 0; nb_tx < num; nb_tx++)
+ rte_pktmbuf_free(pkts_burst[nb_tx]);
+ rte_delay_ms(10);
+ }
+ }
+
+ return ret;
+}
+
+static int
+test_kni_allocate_lcores(void)
+{
+ unsigned i, count = 0;
+
+ lcore_master = rte_get_master_lcore();
+ printf("master lcore: %u\n", lcore_master);
+ for (i = 0; i < RTE_MAX_LCORE; i++) {
+ if (count >=2 )
+ break;
+ if (rte_lcore_is_enabled(i) && i != lcore_master) {
+ count ++;
+ if (count == 1)
+ lcore_ingress = i;
+ else if (count == 2)
+ lcore_egress = i;
+ }
+ }
+ printf("count: %u\n", count);
+
+ return count == 2 ? 0 : -1;
+}
+
+static int
+test_kni_register_handler_mp(void)
+{
+#define TEST_KNI_HANDLE_REQ_COUNT 10 /* 5s */
+#define TEST_KNI_HANDLE_REQ_INTERVAL 500 /* ms */
+#define TEST_KNI_MTU 1450
+#define TEST_KNI_MTU_STR " 1450"
+ int pid;
+
+ pid = fork();
+ if (pid < 0) {
+ printf("Failed to fork a process\n");
+ return -1;
+ } else if (pid == 0) {
+ int i;
+ struct rte_kni *kni = rte_kni_get(TEST_KNI_PORT);
+ struct rte_kni_ops ops = {
+ .change_mtu = kni_change_mtu,
+ .config_network_if = NULL,
+ };
+
+ if (!kni) {
+ printf("Failed to get KNI named %s\n", TEST_KNI_PORT);
+ exit(-1);
+ }
+
+ kni_pkt_mtu = 0;
+
+ /* Check with the invalid parameters */
+ if (rte_kni_register_handlers(kni, NULL) == 0) {
+ printf("Unexpectedly register successuflly "
+ "with NULL ops pointer\n");
+ exit(-1);
+ }
+ if (rte_kni_register_handlers(NULL, &ops) == 0) {
+ printf("Unexpectedly register successfully "
+ "to NULL KNI device pointer\n");
+ exit(-1);
+ }
+
+ if (rte_kni_register_handlers(kni, &ops)) {
+ printf("Fail to register ops\n");
+ exit(-1);
+ }
+
+ /* Check registering again after it has been registered */
+ if (rte_kni_register_handlers(kni, &ops) == 0) {
+ printf("Unexpectedly register successfully after "
+ "it has already been registered\n");
+ exit(-1);
+ }
+
+ /**
+ * Handle the request of setting MTU,
+ * with registered handlers.
+ */
+ for (i = 0; i < TEST_KNI_HANDLE_REQ_COUNT; i++) {
+ rte_kni_handle_request(kni);
+ if (kni_pkt_mtu == TEST_KNI_MTU)
+ break;
+ rte_delay_ms(TEST_KNI_HANDLE_REQ_INTERVAL);
+ }
+ if (i >= TEST_KNI_HANDLE_REQ_COUNT) {
+ printf("MTU has not been set\n");
+ exit(-1);
+ }
+
+ kni_pkt_mtu = 0;
+ if (rte_kni_unregister_handlers(kni) < 0) {
+ printf("Fail to unregister ops\n");
+ exit(-1);
+ }
+
+ /* Check with invalid parameter */
+ if (rte_kni_unregister_handlers(NULL) == 0) {
+ exit(-1);
+ }
+
+ /**
+ * Handle the request of setting MTU,
+ * without registered handlers.
+ */
+ for (i = 0; i < TEST_KNI_HANDLE_REQ_COUNT; i++) {
+ rte_kni_handle_request(kni);
+ if (kni_pkt_mtu != 0)
+ break;
+ rte_delay_ms(TEST_KNI_HANDLE_REQ_INTERVAL);
+ }
+ if (kni_pkt_mtu != 0) {
+ printf("MTU shouldn't be set\n");
+ exit(-1);
+ }
+
+ exit(0);
+ } else {
+ int p_ret, status;
+
+ rte_delay_ms(1000);
+ if (system(IFCONFIG TEST_KNI_PORT " mtu" TEST_KNI_MTU_STR)
+ == -1)
+ return -1;
+
+ rte_delay_ms(1000);
+ if (system(IFCONFIG TEST_KNI_PORT " mtu" TEST_KNI_MTU_STR)
+ == -1)
+ return -1;
+
+ p_ret = wait(&status);
+ if (!WIFEXITED(status)) {
+ printf("Child process (%d) exit abnormally\n", p_ret);
+ return -1;
+ }
+ if (WEXITSTATUS(status) != 0) {
+ printf("Child process exit with failure\n");
+ return -1;
+ }
+ }
+
+ return 0;
+}
+
+static int
+test_kni_processing(uint8_t port_id, struct rte_mempool *mp)
+{
+ int ret = 0;
+ unsigned i;
+ struct rte_kni *kni;
+ struct rte_kni_conf conf;
+ struct rte_eth_dev_info info;
+ struct rte_kni_ops ops;
+
+ if (!mp)
+ return -1;
+
+ memset(&conf, 0, sizeof(conf));
+ memset(&info, 0, sizeof(info));
+ memset(&ops, 0, sizeof(ops));
+
+ rte_eth_dev_info_get(port_id, &info);
+ conf.addr = info.pci_dev->addr;
+ conf.id = info.pci_dev->id;
+ snprintf(conf.name, sizeof(conf.name), TEST_KNI_PORT);
+
+ /* core id 1 configured for kernel thread */
+ conf.core_id = 1;
+ conf.force_bind = 1;
+ conf.mbuf_size = MAX_PACKET_SZ;
+ conf.group_id = (uint16_t)port_id;
+
+ ops = kni_ops;
+ ops.port_id = port_id;
+
+ /* basic test of kni processing */
+ kni = rte_kni_alloc(mp, &conf, &ops);
+ if (!kni) {
+ printf("fail to create kni\n");
+ return -1;
+ }
+
+ test_kni_ctx = kni;
+ test_kni_processing_flag = 0;
+ stats.ingress = 0;
+ stats.egress = 0;
+
+ /**
+ * Check multiple processes support on
+ * registerring/unregisterring handlers.
+ */
+ if (test_kni_register_handler_mp() < 0) {
+ printf("fail to check multiple process support\n");
+ ret = -1;
+ goto fail_kni;
+ }
+
+ rte_eal_mp_remote_launch(test_kni_loop, NULL, CALL_MASTER);
+ RTE_LCORE_FOREACH_SLAVE(i) {
+ if (rte_eal_wait_lcore(i) < 0) {
+ ret = -1;
+ goto fail_kni;
+ }
+ }
+ /**
+ * Check if the number of mbufs received from kernel space is equal
+ * to that of transmitted to kernel space
+ */
+ if (stats.ingress < KNI_NUM_MBUF_THRESHOLD ||
+ stats.egress < KNI_NUM_MBUF_THRESHOLD) {
+ printf("The ingress/egress number should not be "
+ "less than %u\n", (unsigned)KNI_NUM_MBUF_THRESHOLD);
+ ret = -1;
+ goto fail_kni;
+ }
+
+ if (rte_kni_release(kni) < 0) {
+ printf("fail to release kni\n");
+ return -1;
+ }
+ test_kni_ctx = NULL;
+
+ /* test of releasing a released kni device */
+ if (rte_kni_release(kni) == 0) {
+ printf("should not release a released kni device\n");
+ return -1;
+ }
+
+ /* test of reusing memzone */
+ kni = rte_kni_alloc(mp, &conf, &ops);
+ if (!kni) {
+ printf("fail to create kni\n");
+ return -1;
+ }
+
+ /* Release the kni for following testing */
+ if (rte_kni_release(kni) < 0) {
+ printf("fail to release kni\n");
+ return -1;
+ }
+
+ return ret;
+fail_kni:
+ if (rte_kni_release(kni) < 0) {
+ printf("fail to release kni\n");
+ ret = -1;
+ }
+
+ return ret;
+}
+
+static int
+test_kni(void)
+{
+ int ret = -1;
+ uint8_t nb_ports, port_id;
+ struct rte_kni *kni;
+ struct rte_mempool *mp;
+ struct rte_kni_conf conf;
+ struct rte_eth_dev_info info;
+ struct rte_kni_ops ops;
+
+ /* Initialize KNI subsytem */
+ rte_kni_init(KNI_TEST_MAX_PORTS);
+
+ if (test_kni_allocate_lcores() < 0) {
+ printf("No enough lcores for kni processing\n");
+ return -1;
+ }
+
+ mp = test_kni_create_mempool();
+ if (!mp) {
+ printf("fail to create mempool for kni\n");
+ return -1;
+ }
+
+ nb_ports = rte_eth_dev_count();
+ if (nb_ports == 0) {
+ printf("no supported nic port found\n");
+ return -1;
+ }
+
+ /* configuring port 0 for the test is enough */
+ port_id = 0;
+ ret = rte_eth_dev_configure(port_id, 1, 1, &port_conf);
+ if (ret < 0) {
+ printf("fail to configure port %d\n", port_id);
+ return -1;
+ }
+
+ ret = rte_eth_rx_queue_setup(port_id, 0, NB_RXD, SOCKET, &rx_conf, mp);
+ if (ret < 0) {
+ printf("fail to setup rx queue for port %d\n", port_id);
+ return -1;
+ }
+
+ ret = rte_eth_tx_queue_setup(port_id, 0, NB_TXD, SOCKET, &tx_conf);
+ if (ret < 0) {
+ printf("fail to setup tx queue for port %d\n", port_id);
+ return -1;
+ }
+
+ ret = rte_eth_dev_start(port_id);
+ if (ret < 0) {
+ printf("fail to start port %d\n", port_id);
+ return -1;
+ }
+ rte_eth_promiscuous_enable(port_id);
+
+ /* basic test of kni processing */
+ ret = test_kni_processing(port_id, mp);
+ if (ret < 0)
+ goto fail;
+
+ /* test of allocating KNI with NULL mempool pointer */
+ memset(&info, 0, sizeof(info));
+ memset(&conf, 0, sizeof(conf));
+ memset(&ops, 0, sizeof(ops));
+ rte_eth_dev_info_get(port_id, &info);
+ conf.addr = info.pci_dev->addr;
+ conf.id = info.pci_dev->id;
+ conf.group_id = (uint16_t)port_id;
+ conf.mbuf_size = MAX_PACKET_SZ;
+
+ ops = kni_ops;
+ ops.port_id = port_id;
+ kni = rte_kni_alloc(NULL, &conf, &ops);
+ if (kni) {
+ ret = -1;
+ printf("unexpectedly creates kni successfully with NULL "
+ "mempool pointer\n");
+ goto fail;
+ }
+
+ /* test of allocating KNI without configurations */
+ kni = rte_kni_alloc(mp, NULL, NULL);
+ if (kni) {
+ ret = -1;
+ printf("Unexpectedly allocate KNI device successfully "
+ "without configurations\n");
+ goto fail;
+ }
+
+ /* test of allocating KNI without a name */
+ memset(&conf, 0, sizeof(conf));
+ memset(&info, 0, sizeof(info));
+ memset(&ops, 0, sizeof(ops));
+ rte_eth_dev_info_get(port_id, &info);
+ conf.addr = info.pci_dev->addr;
+ conf.id = info.pci_dev->id;
+ conf.group_id = (uint16_t)port_id;
+ conf.mbuf_size = MAX_PACKET_SZ;
+
+ ops = kni_ops;
+ ops.port_id = port_id;
+ kni = rte_kni_alloc(mp, &conf, &ops);
+ if (kni) {
+ ret = -1;
+ printf("Unexpectedly allocate a KNI device successfully "
+ "without a name\n");
+ goto fail;
+ }
+
+ /* test of releasing NULL kni context */
+ ret = rte_kni_release(NULL);
+ if (ret == 0) {
+ ret = -1;
+ printf("unexpectedly release kni successfully\n");
+ goto fail;
+ }
+
+ /* test of handling request on NULL device pointer */
+ ret = rte_kni_handle_request(NULL);
+ if (ret == 0) {
+ ret = -1;
+ printf("Unexpectedly handle request on NULL device pointer\n");
+ goto fail;
+ }
+
+ /* test of getting KNI device with pointer to NULL */
+ kni = rte_kni_get(NULL);
+ if (kni) {
+ ret = -1;
+ printf("Unexpectedly get a KNI device with "
+ "NULL name pointer\n");
+ goto fail;
+ }
+
+ /* test of getting KNI device with an zero length name string */
+ memset(&conf, 0, sizeof(conf));
+ kni = rte_kni_get(conf.name);
+ if (kni) {
+ ret = -1;
+ printf("Unexpectedly get a KNI device with "
+ "zero length name string\n");
+ goto fail;
+ }
+
+ /* test of getting KNI device with an invalid string name */
+ memset(&conf, 0, sizeof(conf));
+ snprintf(conf.name, sizeof(conf.name), "testing");
+ kni = rte_kni_get(conf.name);
+ if (kni) {
+ ret = -1;
+ printf("Unexpectedly get a KNI device with "
+ "a never used name string\n");
+ goto fail;
+ }
+ ret = 0;
+
+fail:
+ rte_eth_dev_stop(port_id);
+
+ return ret;
+}
+
+REGISTER_TEST_COMMAND(kni_autotest, test_kni);
generated by cgit v1.2.3 (git 2.39.1) at 2024-12-18 12:35:46 +0000