* All rights reserved.
*/
+#include <rte_string_fns.h>
#include <rte_mbuf.h>
-#include <rte_ethdev_driver.h>
-#include <rte_ethdev_vdev.h>
+#include <ethdev_driver.h>
+#include <ethdev_vdev.h>
#include <rte_malloc.h>
#include <rte_kvargs.h>
#include <rte_bus_vdev.h>
+#include <errno.h>
#include <linux/if_ether.h>
#include <linux/if_packet.h>
#include <arpa/inet.h>
#include <net/if.h>
+#include <net/if_arp.h>
#include <sys/types.h>
#include <sys/socket.h>
#include <sys/ioctl.h>
+#include <string.h>
#include <sys/mman.h>
#include <unistd.h>
#include <poll.h>
#define ETH_AF_PACKET_FRAMECOUNT_ARG "framecnt"
#define ETH_AF_PACKET_QDISC_BYPASS_ARG "qdisc_bypass"
-#define DFLT_BLOCK_SIZE (1 << 12)
#define DFLT_FRAME_SIZE (1 << 11)
#define DFLT_FRAME_COUNT (1 << 9)
-#define RTE_PMD_AF_PACKET_MAX_RINGS 16
-
struct pkt_rx_queue {
int sockfd;
uint16_t in_port;
volatile unsigned long rx_pkts;
- volatile unsigned long err_pkts;
volatile unsigned long rx_bytes;
};
int if_index;
char *if_name;
- struct ether_addr eth_addr;
+ struct rte_ether_addr eth_addr;
struct tpacket_req req;
- struct pkt_rx_queue rx_queue[RTE_PMD_AF_PACKET_MAX_RINGS];
- struct pkt_tx_queue tx_queue[RTE_PMD_AF_PACKET_MAX_RINGS];
+ struct pkt_rx_queue *rx_queue;
+ struct pkt_tx_queue *tx_queue;
};
static const char *valid_arguments[] = {
.link_autoneg = ETH_LINK_FIXED,
};
-static int af_packet_logtype;
+RTE_LOG_REGISTER_DEFAULT(af_packet_logtype, NOTICE);
#define PMD_LOG(level, fmt, args...) \
rte_log(RTE_LOG_ ## level, af_packet_logtype, \
"%s(): " fmt "\n", __func__, ##args)
+#define PMD_LOG_ERRNO(level, fmt, args...) \
+ rte_log(RTE_LOG_ ## level, af_packet_logtype, \
+ "%s(): " fmt ":%s\n", __func__, ##args, strerror(errno))
+
static uint16_t
eth_af_packet_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts)
{
}
/* kick-off transmits */
- if (sendto(pkt_q->sockfd, NULL, 0, MSG_DONTWAIT, NULL, 0) == -1) {
- /* error sending -- no packets transmitted */
+ if (sendto(pkt_q->sockfd, NULL, 0, MSG_DONTWAIT, NULL, 0) == -1 &&
+ errno != ENOBUFS && errno != EAGAIN) {
+ /*
+ * In case of a ENOBUFS/EAGAIN error all of the enqueued
+ * packets will be considered successful even though only some
+ * are sent.
+ */
+
num_tx = 0;
num_tx_bytes = 0;
}
/*
* This function gets called when the current port gets stopped.
*/
-static void
+static int
eth_dev_stop(struct rte_eth_dev *dev)
{
unsigned i;
}
dev->data->dev_link.link_status = ETH_LINK_DOWN;
+ return 0;
}
static int
return 0;
}
-static void
+static int
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->max_rx_queues = (uint16_t)internals->nb_queues;
dev_info->max_tx_queues = (uint16_t)internals->nb_queues;
dev_info->min_rx_bufsize = 0;
+ dev_info->tx_offload_capa = DEV_TX_OFFLOAD_MULTI_SEGS |
+ DEV_TX_OFFLOAD_VLAN_INSERT;
+
+ return 0;
}
static int
internal->nb_queues : RTE_ETHDEV_QUEUE_STAT_CNTRS);
for (i = 0; i < imax; i++) {
igb_stats->q_opackets[i] = internal->tx_queue[i].tx_pkts;
- igb_stats->q_errors[i] = internal->tx_queue[i].err_pkts;
igb_stats->q_obytes[i] = internal->tx_queue[i].tx_bytes;
tx_total += igb_stats->q_opackets[i];
- tx_err_total += igb_stats->q_errors[i];
+ tx_err_total += internal->tx_queue[i].err_pkts;
tx_bytes_total += igb_stats->q_obytes[i];
}
return 0;
}
-static void
+static int
eth_stats_reset(struct rte_eth_dev *dev)
{
unsigned i;
internal->tx_queue[i].err_pkts = 0;
internal->tx_queue[i].tx_bytes = 0;
}
+
+ return 0;
}
-static void
-eth_dev_close(struct rte_eth_dev *dev __rte_unused)
+static int
+eth_dev_close(struct rte_eth_dev *dev)
{
+ struct pmd_internals *internals;
+ struct tpacket_req *req;
+ unsigned int q;
+
+ if (rte_eal_process_type() != RTE_PROC_PRIMARY)
+ return 0;
+
+ PMD_LOG(INFO, "Closing AF_PACKET ethdev on NUMA socket %u",
+ rte_socket_id());
+
+ internals = dev->data->dev_private;
+ req = &internals->req;
+ for (q = 0; q < internals->nb_queues; q++) {
+ munmap(internals->rx_queue[q].map,
+ 2 * req->tp_block_size * req->tp_block_nr);
+ rte_free(internals->rx_queue[q].rd);
+ rte_free(internals->tx_queue[q].rd);
+ }
+ free(internals->if_name);
+ rte_free(internals->rx_queue);
+ rte_free(internals->tx_queue);
+
+ /* mac_addrs must not be freed alone because part of dev_private */
+ dev->data->mac_addrs = NULL;
+ return 0;
}
static void
int ret;
int s;
unsigned int data_size = internals->req.tp_frame_size -
- TPACKET2_HDRLEN -
- sizeof(struct sockaddr_ll);
+ TPACKET2_HDRLEN;
if (mtu > data_size)
return -EINVAL;
if (s < 0)
return -EINVAL;
- snprintf(ifr.ifr_name, IFNAMSIZ, "%s", internals->if_name);
+ strlcpy(ifr.ifr_name, internals->if_name, IFNAMSIZ);
ret = ioctl(s, SIOCSIFMTU, &ifr);
close(s);
return 0;
}
-static void
+static int
+eth_dev_macaddr_set(struct rte_eth_dev *dev, struct rte_ether_addr *addr)
+{
+ struct pmd_internals *internals = dev->data->dev_private;
+ struct ifreq ifr = { };
+ int sockfd = internals->rx_queue[0].sockfd;
+ int ret;
+
+ if (sockfd == -1) {
+ PMD_LOG(ERR, "receive socket not found");
+ return -EINVAL;
+ }
+
+ strlcpy(ifr.ifr_name, internals->if_name, IFNAMSIZ);
+ ifr.ifr_hwaddr.sa_family = ARPHRD_ETHER;
+ memcpy(ifr.ifr_hwaddr.sa_data, addr, sizeof(*addr));
+ ret = ioctl(sockfd, SIOCSIFHWADDR, &ifr);
+
+ if (ret < 0) {
+ PMD_LOG_ERRNO(ERR, "ioctl(SIOCSIFHWADDR) failed");
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static int
eth_dev_change_flags(char *if_name, uint32_t flags, uint32_t mask)
{
struct ifreq ifr;
+ int ret = 0;
int s;
s = socket(PF_INET, SOCK_DGRAM, 0);
if (s < 0)
- return;
+ return -errno;
- snprintf(ifr.ifr_name, IFNAMSIZ, "%s", if_name);
- if (ioctl(s, SIOCGIFFLAGS, &ifr) < 0)
+ strlcpy(ifr.ifr_name, if_name, IFNAMSIZ);
+ if (ioctl(s, SIOCGIFFLAGS, &ifr) < 0) {
+ ret = -errno;
goto out;
+ }
ifr.ifr_flags &= mask;
ifr.ifr_flags |= flags;
- if (ioctl(s, SIOCSIFFLAGS, &ifr) < 0)
+ if (ioctl(s, SIOCSIFFLAGS, &ifr) < 0) {
+ ret = -errno;
goto out;
+ }
out:
close(s);
+ return ret;
}
-static void
+static int
eth_dev_promiscuous_enable(struct rte_eth_dev *dev)
{
struct pmd_internals *internals = dev->data->dev_private;
- eth_dev_change_flags(internals->if_name, IFF_PROMISC, ~0);
+ return eth_dev_change_flags(internals->if_name, IFF_PROMISC, ~0);
}
-static void
+static int
eth_dev_promiscuous_disable(struct rte_eth_dev *dev)
{
struct pmd_internals *internals = dev->data->dev_private;
- eth_dev_change_flags(internals->if_name, 0, ~IFF_PROMISC);
+ return eth_dev_change_flags(internals->if_name, 0, ~IFF_PROMISC);
}
static const struct eth_dev_ops ops = {
.dev_close = eth_dev_close,
.dev_configure = eth_dev_configure,
.dev_infos_get = eth_dev_info,
+ .mac_addr_set = eth_dev_macaddr_set,
.mtu_set = eth_dev_mtu_set,
.promiscuous_enable = eth_dev_promiscuous_enable,
.promiscuous_disable = eth_dev_promiscuous_disable,
return 0;
}
-static struct rte_vdev_driver pmd_af_packet_drv;
-
static int
rte_pmd_init_internals(struct rte_vdev_device *dev,
const int sockfd,
if (*internals == NULL)
return -1;
+
+ (*internals)->rx_queue = rte_calloc_socket("af_packet_rx",
+ nb_queues,
+ sizeof(struct pkt_rx_queue),
+ 0, numa_node);
+ (*internals)->tx_queue = rte_calloc_socket("af_packet_tx",
+ nb_queues,
+ sizeof(struct pkt_tx_queue),
+ 0, numa_node);
+ if (!(*internals)->rx_queue || !(*internals)->tx_queue) {
+ goto free_internals;
+ }
+
for (q = 0; q < nb_queues; q++) {
(*internals)->rx_queue[q].map = MAP_FAILED;
(*internals)->tx_queue[q].map = MAP_FAILED;
+ (*internals)->rx_queue[q].sockfd = -1;
+ (*internals)->tx_queue[q].sockfd = -1;
}
req = &((*internals)->req);
PMD_LOG(ERR,
"%s: I/F name too long (%s)",
name, pair->value);
- return -1;
+ goto free_internals;
}
if (ioctl(sockfd, SIOCGIFINDEX, &ifr) == -1) {
- PMD_LOG(ERR,
- "%s: ioctl failed (SIOCGIFINDEX)",
- name);
- return -1;
+ PMD_LOG_ERRNO(ERR, "%s: ioctl failed (SIOCGIFINDEX)", name);
+ goto free_internals;
}
(*internals)->if_name = strdup(pair->value);
if ((*internals)->if_name == NULL)
- return -1;
+ goto free_internals;
(*internals)->if_index = ifr.ifr_ifindex;
if (ioctl(sockfd, SIOCGIFHWADDR, &ifr) == -1) {
- PMD_LOG(ERR,
- "%s: ioctl failed (SIOCGIFHWADDR)",
- name);
- return -1;
+ PMD_LOG_ERRNO(ERR, "%s: ioctl failed (SIOCGIFHWADDR)", name);
+ goto free_internals;
}
memcpy(&(*internals)->eth_addr, ifr.ifr_hwaddr.sa_data, ETH_ALEN);
/* Open an AF_PACKET socket for this queue... */
qsockfd = socket(AF_PACKET, SOCK_RAW, htons(ETH_P_ALL));
if (qsockfd == -1) {
- PMD_LOG(ERR,
+ PMD_LOG_ERRNO(ERR,
"%s: could not open AF_PACKET socket",
- name);
- return -1;
+ name);
+ goto error;
}
tpver = TPACKET_V2;
rc = setsockopt(qsockfd, SOL_PACKET, PACKET_VERSION,
&tpver, sizeof(tpver));
if (rc == -1) {
- PMD_LOG(ERR,
+ PMD_LOG_ERRNO(ERR,
"%s: could not set PACKET_VERSION on AF_PACKET socket for %s",
name, pair->value);
goto error;
rc = setsockopt(qsockfd, SOL_PACKET, PACKET_LOSS,
&discard, sizeof(discard));
if (rc == -1) {
- PMD_LOG(ERR,
+ PMD_LOG_ERRNO(ERR,
"%s: could not set PACKET_LOSS on AF_PACKET socket for %s",
name, pair->value);
goto error;
rc = setsockopt(qsockfd, SOL_PACKET, PACKET_QDISC_BYPASS,
&qdisc_bypass, sizeof(qdisc_bypass));
if (rc == -1) {
- PMD_LOG(ERR,
+ PMD_LOG_ERRNO(ERR,
"%s: could not set PACKET_QDISC_BYPASS on AF_PACKET socket for %s",
name, pair->value);
goto error;
rc = setsockopt(qsockfd, SOL_PACKET, PACKET_RX_RING, req, sizeof(*req));
if (rc == -1) {
- PMD_LOG(ERR,
+ PMD_LOG_ERRNO(ERR,
"%s: could not set PACKET_RX_RING on AF_PACKET socket for %s",
name, pair->value);
goto error;
rc = setsockopt(qsockfd, SOL_PACKET, PACKET_TX_RING, req, sizeof(*req));
if (rc == -1) {
- PMD_LOG(ERR,
+ PMD_LOG_ERRNO(ERR,
"%s: could not set PACKET_TX_RING on AF_PACKET "
"socket for %s", name, pair->value);
goto error;
PROT_READ | PROT_WRITE, MAP_SHARED | MAP_LOCKED,
qsockfd, 0);
if (rx_queue->map == MAP_FAILED) {
- PMD_LOG(ERR,
+ PMD_LOG_ERRNO(ERR,
"%s: call to mmap failed on AF_PACKET socket for %s",
name, pair->value);
goto error;
rc = bind(qsockfd, (const struct sockaddr*)&sockaddr, sizeof(sockaddr));
if (rc == -1) {
- PMD_LOG(ERR,
+ PMD_LOG_ERRNO(ERR,
"%s: could not bind AF_PACKET socket to %s",
- name, pair->value);
+ name, pair->value);
goto error;
}
rc = setsockopt(qsockfd, SOL_PACKET, PACKET_FANOUT,
&fanout_arg, sizeof(fanout_arg));
if (rc == -1) {
- PMD_LOG(ERR,
- "%s: could not set PACKET_FANOUT on AF_PACKET socket "
- "for %s", name, pair->value);
+ PMD_LOG_ERRNO(ERR,
+ "%s: could not set PACKET_FANOUT on AF_PACKET socket for %s",
+ name, pair->value);
goto error;
}
#endif
data->nb_tx_queues = (uint16_t)nb_queues;
data->dev_link = pmd_link;
data->mac_addrs = &(*internals)->eth_addr;
+ data->dev_flags |= RTE_ETH_DEV_AUTOFILL_QUEUE_XSTATS;
(*eth_dev)->dev_ops = &ops;
if (qsockfd != -1)
close(qsockfd);
for (q = 0; q < nb_queues; q++) {
- munmap((*internals)->rx_queue[q].map,
- 2 * req->tp_block_size * req->tp_block_nr);
+ if ((*internals)->rx_queue[q].map != MAP_FAILED)
+ munmap((*internals)->rx_queue[q].map,
+ 2 * req->tp_block_size * req->tp_block_nr);
rte_free((*internals)->rx_queue[q].rd);
rte_free((*internals)->tx_queue[q].rd);
- if (((*internals)->rx_queue[q].sockfd != 0) &&
+ if (((*internals)->rx_queue[q].sockfd >= 0) &&
((*internals)->rx_queue[q].sockfd != qsockfd))
close((*internals)->rx_queue[q].sockfd);
}
+free_internals:
+ rte_free((*internals)->rx_queue);
+ rte_free((*internals)->tx_queue);
free((*internals)->if_name);
rte_free(*internals);
return -1;
struct rte_kvargs_pair *pair = NULL;
unsigned k_idx;
unsigned int blockcount;
- unsigned int blocksize = DFLT_BLOCK_SIZE;
+ unsigned int blocksize;
unsigned int framesize = DFLT_FRAME_SIZE;
unsigned int framecount = DFLT_FRAME_COUNT;
unsigned int qpairs = 1;
if (*sockfd < 0)
return -1;
+ blocksize = getpagesize();
+
/*
* Walk arguments for configurable settings
*/
pair = &kvlist->pairs[k_idx];
if (strstr(pair->key, ETH_AF_PACKET_NUM_Q_ARG) != NULL) {
qpairs = atoi(pair->value);
- if (qpairs < 1 ||
- qpairs > RTE_PMD_AF_PACKET_MAX_RINGS) {
+ if (qpairs < 1) {
PMD_LOG(ERR,
"%s: invalid qpairs value",
name);
eth_dev->rx_pkt_burst = eth_af_packet_rx;
eth_dev->tx_pkt_burst = eth_af_packet_tx;
+ rte_eth_dev_probing_finish(eth_dev);
return 0;
}
PMD_LOG(INFO, "Initializing pmd_af_packet for %s", name);
- if (rte_eal_process_type() == RTE_PROC_SECONDARY &&
- strlen(rte_vdev_device_args(dev)) == 0) {
+ if (rte_eal_process_type() == RTE_PROC_SECONDARY) {
eth_dev = rte_eth_dev_attach_secondary(name);
if (!eth_dev) {
PMD_LOG(ERR, "Failed to probe %s", name);
}
/* TODO: request info from primary to set up Rx and Tx */
eth_dev->dev_ops = &ops;
+ eth_dev->device = &dev->device;
+ rte_eth_dev_probing_finish(eth_dev);
return 0;
}
static int
rte_pmd_af_packet_remove(struct rte_vdev_device *dev)
{
- struct rte_eth_dev *eth_dev = NULL;
- struct pmd_internals *internals;
- unsigned q;
-
- PMD_LOG(INFO, "Closing AF_PACKET ethdev on numa socket %u",
- rte_socket_id());
+ struct rte_eth_dev *eth_dev;
if (dev == NULL)
return -1;
/* find the ethdev entry */
eth_dev = rte_eth_dev_allocated(rte_vdev_device_name(dev));
if (eth_dev == NULL)
- return -1;
-
- internals = eth_dev->data->dev_private;
- for (q = 0; q < internals->nb_queues; q++) {
- rte_free(internals->rx_queue[q].rd);
- rte_free(internals->tx_queue[q].rd);
- }
- free(internals->if_name);
-
- rte_free(eth_dev->data->dev_private);
+ return 0; /* port already released */
+ eth_dev_close(eth_dev);
rte_eth_dev_release_port(eth_dev);
return 0;
"framesz=<int> "
"framecnt=<int> "
"qdisc_bypass=<0|1>");
-
-RTE_INIT(af_packet_init_log);
-static void
-af_packet_init_log(void)
-{
- af_packet_logtype = rte_log_register("pmd.net.packet");
- if (af_packet_logtype >= 0)
- rte_log_set_level(af_packet_logtype, RTE_LOG_NOTICE);
-}