X-Git-Url: http://git.droids-corp.org/?a=blobdiff_plain;f=drivers%2Fnet%2Fpcap%2Frte_eth_pcap.c;h=4e6d49370eb345e17e5e3cf9c26ce7f2cdff5c9f;hb=d23d73d088c1;hp=595d4f8c05f7e95aff34a565a34fc721a8a778f1;hpb=3c8870f8ff2f0f9c7cd00eb2fcc3f7221ef7be13;p=dpdk.git diff --git a/drivers/net/pcap/rte_eth_pcap.c b/drivers/net/pcap/rte_eth_pcap.c index 595d4f8c05..4e6d49370e 100644 --- a/drivers/net/pcap/rte_eth_pcap.c +++ b/drivers/net/pcap/rte_eth_pcap.c @@ -1,146 +1,156 @@ -/*- - * BSD LICENSE - * - * Copyright(c) 2010-2016 Intel Corporation. All rights reserved. - * Copyright(c) 2014 6WIND S.A. - * 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. +/* SPDX-License-Identifier: BSD-3-Clause + * Copyright(c) 2010-2016 Intel Corporation. + * Copyright(c) 2014 6WIND S.A. + * All rights reserved. */ #include -#include -#include -#include -#include -#include -#include -#include -#include #include +#include +#include +#include + +#if defined(RTE_EXEC_ENV_FREEBSD) +#include +#include +#endif #include +#include +#include +#include +#include +#include +#include +#include +#include + #define RTE_ETH_PCAP_SNAPSHOT_LEN 65535 -#define RTE_ETH_PCAP_SNAPLEN ETHER_MAX_JUMBO_FRAME_LEN +#define RTE_ETH_PCAP_SNAPLEN RTE_ETHER_MAX_JUMBO_FRAME_LEN #define RTE_ETH_PCAP_PROMISC 1 #define RTE_ETH_PCAP_TIMEOUT -1 + #define ETH_PCAP_RX_PCAP_ARG "rx_pcap" #define ETH_PCAP_TX_PCAP_ARG "tx_pcap" #define ETH_PCAP_RX_IFACE_ARG "rx_iface" +#define ETH_PCAP_RX_IFACE_IN_ARG "rx_iface_in" #define ETH_PCAP_TX_IFACE_ARG "tx_iface" #define ETH_PCAP_IFACE_ARG "iface" +#define ETH_PCAP_PHY_MAC_ARG "phy_mac" +#define ETH_PCAP_INFINITE_RX_ARG "infinite_rx" #define ETH_PCAP_ARG_MAXLEN 64 #define RTE_PMD_PCAP_MAX_QUEUES 16 static char errbuf[PCAP_ERRBUF_SIZE]; -static unsigned char tx_pcap_data[RTE_ETH_PCAP_SNAPLEN]; static struct timeval start_time; static uint64_t start_cycles; static uint64_t hz; +static uint8_t iface_idx; + +static uint64_t timestamp_rx_dynflag; +static int timestamp_dynfield_offset = -1; + +struct queue_stat { + volatile unsigned long pkts; + volatile unsigned long bytes; + volatile unsigned long err_pkts; +}; struct pcap_rx_queue { - pcap_t *pcap; - uint8_t in_port; + uint16_t port_id; + uint16_t queue_id; struct rte_mempool *mb_pool; - volatile unsigned long rx_pkts; - volatile unsigned long rx_bytes; - volatile unsigned long err_pkts; + struct queue_stat rx_stat; char name[PATH_MAX]; char type[ETH_PCAP_ARG_MAXLEN]; + + /* Contains pre-generated packets to be looped through */ + struct rte_ring *pkts; }; struct pcap_tx_queue { - pcap_dumper_t *dumper; - pcap_t *pcap; - volatile unsigned long tx_pkts; - volatile unsigned long tx_bytes; - volatile unsigned long err_pkts; + uint16_t port_id; + uint16_t queue_id; + struct queue_stat tx_stat; char name[PATH_MAX]; char type[ETH_PCAP_ARG_MAXLEN]; }; -struct rx_pcaps { - unsigned num_of_rx; - pcap_t *pcaps[RTE_PMD_PCAP_MAX_QUEUES]; - const char *names[RTE_PMD_PCAP_MAX_QUEUES]; - const char *types[RTE_PMD_PCAP_MAX_QUEUES]; -}; - -struct tx_pcaps { - unsigned num_of_tx; - pcap_dumper_t *dumpers[RTE_PMD_PCAP_MAX_QUEUES]; - pcap_t *pcaps[RTE_PMD_PCAP_MAX_QUEUES]; - const char *names[RTE_PMD_PCAP_MAX_QUEUES]; - const char *types[RTE_PMD_PCAP_MAX_QUEUES]; -}; - struct pmd_internals { struct pcap_rx_queue rx_queue[RTE_PMD_PCAP_MAX_QUEUES]; struct pcap_tx_queue tx_queue[RTE_PMD_PCAP_MAX_QUEUES]; + char devargs[ETH_PCAP_ARG_MAXLEN]; + struct rte_ether_addr eth_addr; int if_index; int single_iface; + int phy_mac; + unsigned int infinite_rx; +}; + +struct pmd_process_private { + pcap_t *rx_pcap[RTE_PMD_PCAP_MAX_QUEUES]; + pcap_t *tx_pcap[RTE_PMD_PCAP_MAX_QUEUES]; + pcap_dumper_t *tx_dumper[RTE_PMD_PCAP_MAX_QUEUES]; +}; + +struct pmd_devargs { + unsigned int num_of_queue; + struct devargs_queue { + pcap_dumper_t *dumper; + pcap_t *pcap; + const char *name; + const char *type; + } queue[RTE_PMD_PCAP_MAX_QUEUES]; + int phy_mac; +}; + +struct pmd_devargs_all { + struct pmd_devargs rx_queues; + struct pmd_devargs tx_queues; + int single_iface; + unsigned int is_tx_pcap; + unsigned int is_tx_iface; + unsigned int is_rx_pcap; + unsigned int is_rx_iface; + unsigned int infinite_rx; }; -const char *valid_arguments[] = { +static const char *valid_arguments[] = { ETH_PCAP_RX_PCAP_ARG, ETH_PCAP_TX_PCAP_ARG, ETH_PCAP_RX_IFACE_ARG, + ETH_PCAP_RX_IFACE_IN_ARG, ETH_PCAP_TX_IFACE_ARG, ETH_PCAP_IFACE_ARG, + ETH_PCAP_PHY_MAC_ARG, + ETH_PCAP_INFINITE_RX_ARG, NULL }; -static int open_single_tx_pcap(const char *pcap_filename, pcap_dumper_t **dumper); -static int open_single_rx_pcap(const char *pcap_filename, pcap_t **pcap); -static int open_single_iface(const char *iface, pcap_t **pcap); - -static struct ether_addr eth_addr = { .addr_bytes = { 0, 0, 0, 0x1, 0x2, 0x3 } }; -static const char *drivername = "Pcap PMD"; static struct rte_eth_link pmd_link = { .link_speed = ETH_SPEED_NUM_10G, .link_duplex = ETH_LINK_FULL_DUPLEX, .link_status = ETH_LINK_DOWN, - .link_autoneg = ETH_LINK_SPEED_FIXED, + .link_autoneg = ETH_LINK_FIXED, }; +RTE_LOG_REGISTER(eth_pcap_logtype, pmd.net.pcap, NOTICE); + +#define PMD_LOG(level, fmt, args...) \ + rte_log(RTE_LOG_ ## level, eth_pcap_logtype, \ + "%s(): " fmt "\n", __func__, ##args) + static int -eth_pcap_rx_jumbo(struct rte_mempool *mb_pool, - struct rte_mbuf *mbuf, - const u_char *data, - uint16_t data_len) +eth_pcap_rx_jumbo(struct rte_mempool *mb_pool, struct rte_mbuf *mbuf, + const u_char *data, uint16_t data_len) { - struct rte_mbuf *m = mbuf; - /* Copy the first segment. */ uint16_t len = rte_pktmbuf_tailroom(mbuf); + struct rte_mbuf *m = mbuf; rte_memcpy(rte_pktmbuf_append(mbuf, len), data, len); data_len -= len; @@ -172,36 +182,60 @@ eth_pcap_rx_jumbo(struct rte_mempool *mb_pool, return mbuf->nb_segs; } -/* Copy data from mbuf chain to a buffer suitable for writing to a PCAP file. */ -static void -eth_pcap_gather_data(unsigned char *data, struct rte_mbuf *mbuf) +static uint16_t +eth_pcap_rx_infinite(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) { - uint16_t data_len = 0; + int i; + struct pcap_rx_queue *pcap_q = queue; + uint32_t rx_bytes = 0; - while (mbuf) { - rte_memcpy(data + data_len, rte_pktmbuf_mtod(mbuf, void *), - mbuf->data_len); + if (unlikely(nb_pkts == 0)) + return 0; - data_len += mbuf->data_len; - mbuf = mbuf->next; + if (rte_pktmbuf_alloc_bulk(pcap_q->mb_pool, bufs, nb_pkts) != 0) + return 0; + + for (i = 0; i < nb_pkts; i++) { + struct rte_mbuf *pcap_buf; + int err = rte_ring_dequeue(pcap_q->pkts, (void **)&pcap_buf); + if (err) + return i; + + rte_memcpy(rte_pktmbuf_mtod(bufs[i], void *), + rte_pktmbuf_mtod(pcap_buf, void *), + pcap_buf->data_len); + bufs[i]->data_len = pcap_buf->data_len; + bufs[i]->pkt_len = pcap_buf->pkt_len; + bufs[i]->port = pcap_q->port_id; + rx_bytes += pcap_buf->data_len; + + /* Enqueue packet back on ring to allow infinite rx. */ + rte_ring_enqueue(pcap_q->pkts, pcap_buf); } + + pcap_q->rx_stat.pkts += i; + pcap_q->rx_stat.bytes += rx_bytes; + + return i; } static uint16_t -eth_pcap_rx(void *queue, - struct rte_mbuf **bufs, - uint16_t nb_pkts) +eth_pcap_rx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) { - unsigned i; + unsigned int i; struct pcap_pkthdr header; + struct pmd_process_private *pp; const u_char *packet; struct rte_mbuf *mbuf; struct pcap_rx_queue *pcap_q = queue; uint16_t num_rx = 0; - uint16_t buf_size; uint32_t rx_bytes = 0; + pcap_t *pcap; - if (unlikely(pcap_q->pcap == NULL || nb_pkts == 0)) + pp = rte_eth_devices[pcap_q->port_id].process_private; + pcap = pp->rx_pcap[pcap_q->queue_id]; + + if (unlikely(pcap == NULL || nb_pkts == 0)) return 0; /* Reads the given number of packets from the pcap file one by one @@ -209,20 +243,16 @@ eth_pcap_rx(void *queue, */ for (i = 0; i < nb_pkts; i++) { /* Get the next PCAP packet */ - packet = pcap_next(pcap_q->pcap, &header); + packet = pcap_next(pcap, &header); if (unlikely(packet == NULL)) break; - else - mbuf = rte_pktmbuf_alloc(pcap_q->mb_pool); + + mbuf = rte_pktmbuf_alloc(pcap_q->mb_pool); if (unlikely(mbuf == NULL)) break; - /* Now get the space available for data in the mbuf */ - buf_size = (uint16_t)(rte_pktmbuf_data_room_size(pcap_q->mb_pool) - - RTE_PKTMBUF_HEADROOM); - - if (header.caplen <= buf_size) { - /* pcap packet will fit in the mbuf, go ahead and copy */ + if (header.caplen <= rte_pktmbuf_tailroom(mbuf)) { + /* pcap packet will fit in the mbuf, can copy it */ rte_memcpy(rte_pktmbuf_mtod(mbuf, void *), packet, header.caplen); mbuf->data_len = (uint16_t)header.caplen; @@ -231,21 +261,39 @@ eth_pcap_rx(void *queue, if (unlikely(eth_pcap_rx_jumbo(pcap_q->mb_pool, mbuf, packet, - header.caplen) == -1)) + header.caplen) == -1)) { + rte_pktmbuf_free(mbuf); break; + } } mbuf->pkt_len = (uint16_t)header.caplen; - mbuf->port = pcap_q->in_port; + *RTE_MBUF_DYNFIELD(mbuf, timestamp_dynfield_offset, + rte_mbuf_timestamp_t *) = + (uint64_t)header.ts.tv_sec * 1000000 + + header.ts.tv_usec; + mbuf->ol_flags |= timestamp_rx_dynflag; + mbuf->port = pcap_q->port_id; bufs[num_rx] = mbuf; num_rx++; rx_bytes += header.caplen; } - pcap_q->rx_pkts += num_rx; - pcap_q->rx_bytes += rx_bytes; + pcap_q->rx_stat.pkts += num_rx; + pcap_q->rx_stat.bytes += rx_bytes; + return num_rx; } +static uint16_t +eth_null_rx(void *queue __rte_unused, + struct rte_mbuf **bufs __rte_unused, + uint16_t nb_pkts __rte_unused) +{ + return 0; +} + +#define NSEC_PER_SEC 1000000000L + static inline void calculate_timestamp(struct timeval *ts) { uint64_t cycles; @@ -253,58 +301,62 @@ calculate_timestamp(struct timeval *ts) { cycles = rte_get_timer_cycles() - start_cycles; cur_time.tv_sec = cycles / hz; - cur_time.tv_usec = (cycles % hz) * 10e6 / hz; - timeradd(&start_time, &cur_time, ts); + cur_time.tv_usec = (cycles % hz) * NSEC_PER_SEC / hz; + + ts->tv_sec = start_time.tv_sec + cur_time.tv_sec; + ts->tv_usec = start_time.tv_usec + cur_time.tv_usec; + if (ts->tv_usec >= NSEC_PER_SEC) { + ts->tv_usec -= NSEC_PER_SEC; + ts->tv_sec += 1; + } } /* * Callback to handle writing packets to a pcap file. */ static uint16_t -eth_pcap_tx_dumper(void *queue, - struct rte_mbuf **bufs, - uint16_t nb_pkts) +eth_pcap_tx_dumper(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) { - unsigned i; + unsigned int i; struct rte_mbuf *mbuf; + struct pmd_process_private *pp; struct pcap_tx_queue *dumper_q = queue; uint16_t num_tx = 0; uint32_t tx_bytes = 0; struct pcap_pkthdr header; + pcap_dumper_t *dumper; + unsigned char temp_data[RTE_ETH_PCAP_SNAPLEN]; + size_t len, caplen; - if (dumper_q->dumper == NULL || nb_pkts == 0) + pp = rte_eth_devices[dumper_q->port_id].process_private; + dumper = pp->tx_dumper[dumper_q->queue_id]; + + if (dumper == NULL || nb_pkts == 0) return 0; - /* writes the nb_pkts packets to the previously opened pcap file dumper */ + /* writes the nb_pkts packets to the previously opened pcap file + * dumper */ for (i = 0; i < nb_pkts; i++) { mbuf = bufs[i]; - calculate_timestamp(&header.ts); - header.len = mbuf->pkt_len; - header.caplen = header.len; - - if (likely(mbuf->nb_segs == 1)) { - pcap_dump((u_char *)dumper_q->dumper, &header, - rte_pktmbuf_mtod(mbuf, void*)); - } else { - if (mbuf->pkt_len <= ETHER_MAX_JUMBO_FRAME_LEN) { - eth_pcap_gather_data(tx_pcap_data, mbuf); - pcap_dump((u_char *)dumper_q->dumper, &header, - tx_pcap_data); - } else { - RTE_LOG(ERR, PMD, - "Dropping PCAP packet. " - "Size (%d) > max jumbo size (%d).\n", - mbuf->pkt_len, - ETHER_MAX_JUMBO_FRAME_LEN); - - rte_pktmbuf_free(mbuf); - break; - } + len = caplen = rte_pktmbuf_pkt_len(mbuf); + if (unlikely(!rte_pktmbuf_is_contiguous(mbuf) && + len > sizeof(temp_data))) { + caplen = sizeof(temp_data); } - rte_pktmbuf_free(mbuf); + calculate_timestamp(&header.ts); + header.len = len; + header.caplen = caplen; + /* rte_pktmbuf_read() returns a pointer to the data directly + * in the mbuf (when the mbuf is contiguous) or, otherwise, + * a pointer to temp_data after copying into it. + */ + pcap_dump((u_char *)dumper, &header, + rte_pktmbuf_read(mbuf, 0, caplen, temp_data)); + num_tx++; - tx_bytes += mbuf->pkt_len; + tx_bytes += caplen; + rte_pktmbuf_free(mbuf); } /* @@ -312,74 +364,187 @@ eth_pcap_tx_dumper(void *queue, * process stops and to make sure the pcap file is actually written, * we flush the pcap dumper within each burst. */ - pcap_dump_flush(dumper_q->dumper); - dumper_q->tx_pkts += num_tx; - dumper_q->tx_bytes += tx_bytes; - dumper_q->err_pkts += nb_pkts - num_tx; - return num_tx; + pcap_dump_flush(dumper); + dumper_q->tx_stat.pkts += num_tx; + dumper_q->tx_stat.bytes += tx_bytes; + dumper_q->tx_stat.err_pkts += nb_pkts - num_tx; + + return nb_pkts; +} + +/* + * Callback to handle dropping packets in the infinite rx case. + */ +static uint16_t +eth_tx_drop(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) +{ + unsigned int i; + uint32_t tx_bytes = 0; + struct pcap_tx_queue *tx_queue = queue; + + if (unlikely(nb_pkts == 0)) + return 0; + + for (i = 0; i < nb_pkts; i++) { + tx_bytes += bufs[i]->data_len; + rte_pktmbuf_free(bufs[i]); + } + + tx_queue->tx_stat.pkts += nb_pkts; + tx_queue->tx_stat.bytes += tx_bytes; + + return i; } /* * Callback to handle sending packets through a real NIC. */ static uint16_t -eth_pcap_tx(void *queue, - struct rte_mbuf **bufs, - uint16_t nb_pkts) +eth_pcap_tx(void *queue, struct rte_mbuf **bufs, uint16_t nb_pkts) { - unsigned i; + unsigned int i; int ret; struct rte_mbuf *mbuf; + struct pmd_process_private *pp; struct pcap_tx_queue *tx_queue = queue; uint16_t num_tx = 0; uint32_t tx_bytes = 0; + pcap_t *pcap; + unsigned char temp_data[RTE_ETH_PCAP_SNAPLEN]; + size_t len; + + pp = rte_eth_devices[tx_queue->port_id].process_private; + pcap = pp->tx_pcap[tx_queue->queue_id]; - if (unlikely(nb_pkts == 0 || tx_queue->pcap == NULL)) + if (unlikely(nb_pkts == 0 || pcap == NULL)) return 0; for (i = 0; i < nb_pkts; i++) { mbuf = bufs[i]; - - if (likely(mbuf->nb_segs == 1)) { - ret = pcap_sendpacket(tx_queue->pcap, - rte_pktmbuf_mtod(mbuf, u_char *), - mbuf->pkt_len); - } else { - if (mbuf->pkt_len <= ETHER_MAX_JUMBO_FRAME_LEN) { - eth_pcap_gather_data(tx_pcap_data, mbuf); - ret = pcap_sendpacket(tx_queue->pcap, - tx_pcap_data, - mbuf->pkt_len); - } else { - RTE_LOG(ERR, PMD, - "Dropping PCAP packet. " - "Size (%d) > max jumbo size (%d).\n", - mbuf->pkt_len, - ETHER_MAX_JUMBO_FRAME_LEN); - - rte_pktmbuf_free(mbuf); - break; - } + len = rte_pktmbuf_pkt_len(mbuf); + if (unlikely(!rte_pktmbuf_is_contiguous(mbuf) && + len > sizeof(temp_data))) { + PMD_LOG(ERR, + "Dropping multi segment PCAP packet. Size (%zd) > max size (%zd).", + len, sizeof(temp_data)); + rte_pktmbuf_free(mbuf); + continue; } + /* rte_pktmbuf_read() returns a pointer to the data directly + * in the mbuf (when the mbuf is contiguous) or, otherwise, + * a pointer to temp_data after copying into it. + */ + ret = pcap_sendpacket(pcap, + rte_pktmbuf_read(mbuf, 0, len, temp_data), len); if (unlikely(ret != 0)) break; num_tx++; - tx_bytes += mbuf->pkt_len; + tx_bytes += len; rte_pktmbuf_free(mbuf); } - tx_queue->tx_pkts += num_tx; - tx_queue->tx_bytes += tx_bytes; - tx_queue->err_pkts += nb_pkts - num_tx; - return num_tx; + tx_queue->tx_stat.pkts += num_tx; + tx_queue->tx_stat.bytes += tx_bytes; + tx_queue->tx_stat.err_pkts += i - num_tx; + + return i; +} + +/* + * pcap_open_live wrapper function + */ +static inline int +open_iface_live(const char *iface, pcap_t **pcap) { + *pcap = pcap_open_live(iface, RTE_ETH_PCAP_SNAPLEN, + RTE_ETH_PCAP_PROMISC, RTE_ETH_PCAP_TIMEOUT, errbuf); + + if (*pcap == NULL) { + PMD_LOG(ERR, "Couldn't open %s: %s", iface, errbuf); + return -1; + } + + return 0; +} + +static int +open_single_iface(const char *iface, pcap_t **pcap) +{ + if (open_iface_live(iface, pcap) < 0) { + PMD_LOG(ERR, "Couldn't open interface %s", iface); + return -1; + } + + return 0; +} + +static int +open_single_tx_pcap(const char *pcap_filename, pcap_dumper_t **dumper) +{ + pcap_t *tx_pcap; + + /* + * We need to create a dummy empty pcap_t to use it + * with pcap_dump_open(). We create big enough an Ethernet + * pcap holder. + */ + tx_pcap = pcap_open_dead_with_tstamp_precision(DLT_EN10MB, + RTE_ETH_PCAP_SNAPSHOT_LEN, PCAP_TSTAMP_PRECISION_NANO); + if (tx_pcap == NULL) { + PMD_LOG(ERR, "Couldn't create dead pcap"); + return -1; + } + + /* The dumper is created using the previous pcap_t reference */ + *dumper = pcap_dump_open(tx_pcap, pcap_filename); + if (*dumper == NULL) { + pcap_close(tx_pcap); + PMD_LOG(ERR, "Couldn't open %s for writing.", + pcap_filename); + return -1; + } + + pcap_close(tx_pcap); + return 0; +} + +static int +open_single_rx_pcap(const char *pcap_filename, pcap_t **pcap) +{ + *pcap = pcap_open_offline(pcap_filename, errbuf); + if (*pcap == NULL) { + PMD_LOG(ERR, "Couldn't open %s: %s", pcap_filename, + errbuf); + return -1; + } + + return 0; +} + +static uint64_t +count_packets_in_pcap(pcap_t **pcap, struct pcap_rx_queue *pcap_q) +{ + const u_char *packet; + struct pcap_pkthdr header; + uint64_t pcap_pkt_count = 0; + + while ((packet = pcap_next(*pcap, &header))) + pcap_pkt_count++; + + /* The pcap is reopened so it can be used as normal later. */ + pcap_close(*pcap); + *pcap = NULL; + open_single_rx_pcap(pcap_q->name, pcap); + + return pcap_pkt_count; } static int eth_dev_start(struct rte_eth_dev *dev) { - unsigned i; + unsigned int i; struct pmd_internals *internals = dev->data->dev_private; + struct pmd_process_private *pp = dev->process_private; struct pcap_tx_queue *tx; struct pcap_rx_queue *rx; @@ -388,11 +553,13 @@ eth_dev_start(struct rte_eth_dev *dev) tx = &internals->tx_queue[0]; rx = &internals->rx_queue[0]; - if (!tx->pcap && strcmp(tx->type, ETH_PCAP_IFACE_ARG) == 0) { - if (open_single_iface(tx->name, &tx->pcap) < 0) + if (!pp->tx_pcap[0] && + strcmp(tx->type, ETH_PCAP_IFACE_ARG) == 0) { + if (open_single_iface(tx->name, &pp->tx_pcap[0]) < 0) return -1; - rx->pcap = tx->pcap; + pp->rx_pcap[0] = pp->tx_pcap[0]; } + goto status_up; } @@ -400,13 +567,14 @@ eth_dev_start(struct rte_eth_dev *dev) for (i = 0; i < dev->data->nb_tx_queues; i++) { tx = &internals->tx_queue[i]; - if (!tx->dumper && strcmp(tx->type, ETH_PCAP_TX_PCAP_ARG) == 0) { - if (open_single_tx_pcap(tx->name, &tx->dumper) < 0) + if (!pp->tx_dumper[i] && + strcmp(tx->type, ETH_PCAP_TX_PCAP_ARG) == 0) { + if (open_single_tx_pcap(tx->name, + &pp->tx_dumper[i]) < 0) return -1; - } - - else if (!tx->pcap && strcmp(tx->type, ETH_PCAP_TX_IFACE_ARG) == 0) { - if (open_single_iface(tx->name, &tx->pcap) < 0) + } else if (!pp->tx_pcap[i] && + strcmp(tx->type, ETH_PCAP_TX_IFACE_ARG) == 0) { + if (open_single_iface(tx->name, &pp->tx_pcap[i]) < 0) return -1; } } @@ -415,23 +583,27 @@ eth_dev_start(struct rte_eth_dev *dev) for (i = 0; i < dev->data->nb_rx_queues; i++) { rx = &internals->rx_queue[i]; - if (rx->pcap != NULL) + if (pp->rx_pcap[i] != NULL) continue; if (strcmp(rx->type, ETH_PCAP_RX_PCAP_ARG) == 0) { - if (open_single_rx_pcap(rx->name, &rx->pcap) < 0) + if (open_single_rx_pcap(rx->name, &pp->rx_pcap[i]) < 0) return -1; - } - - else if (strcmp(rx->type, ETH_PCAP_RX_IFACE_ARG) == 0) { - if (open_single_iface(rx->name, &rx->pcap) < 0) + } else if (strcmp(rx->type, ETH_PCAP_RX_IFACE_ARG) == 0) { + if (open_single_iface(rx->name, &pp->rx_pcap[i]) < 0) return -1; } } status_up: + for (i = 0; i < dev->data->nb_rx_queues; i++) + dev->data->rx_queue_state[i] = RTE_ETH_QUEUE_STATE_STARTED; + + for (i = 0; i < dev->data->nb_tx_queues; i++) + dev->data->tx_queue_state[i] = RTE_ETH_QUEUE_STATE_STARTED; dev->data->dev_link.link_status = ETH_LINK_UP; + return 0; } @@ -440,77 +612,87 @@ status_up: * Is the only place for us to close all the tx streams dumpers. * If not called the dumpers will be flushed within each tx burst. */ -static void +static int eth_dev_stop(struct rte_eth_dev *dev) { - unsigned i; + unsigned int i; struct pmd_internals *internals = dev->data->dev_private; - struct pcap_tx_queue *tx; - struct pcap_rx_queue *rx; + struct pmd_process_private *pp = dev->process_private; /* Special iface case. Single pcap is open and shared between tx/rx. */ if (internals->single_iface) { - tx = &internals->tx_queue[0]; - rx = &internals->rx_queue[0]; - pcap_close(tx->pcap); - tx->pcap = NULL; - rx->pcap = NULL; + pcap_close(pp->tx_pcap[0]); + pp->tx_pcap[0] = NULL; + pp->rx_pcap[0] = NULL; goto status_down; } for (i = 0; i < dev->data->nb_tx_queues; i++) { - tx = &internals->tx_queue[i]; - - if (tx->dumper != NULL) { - pcap_dump_close(tx->dumper); - tx->dumper = NULL; + if (pp->tx_dumper[i] != NULL) { + pcap_dump_close(pp->tx_dumper[i]); + pp->tx_dumper[i] = NULL; } - if (tx->pcap != NULL) { - pcap_close(tx->pcap); - tx->pcap = NULL; + if (pp->tx_pcap[i] != NULL) { + pcap_close(pp->tx_pcap[i]); + pp->tx_pcap[i] = NULL; } } for (i = 0; i < dev->data->nb_rx_queues; i++) { - rx = &internals->rx_queue[i]; - - if (rx->pcap != NULL) { - pcap_close(rx->pcap); - rx->pcap = NULL; + if (pp->rx_pcap[i] != NULL) { + pcap_close(pp->rx_pcap[i]); + pp->rx_pcap[i] = NULL; } } status_down: + for (i = 0; i < dev->data->nb_rx_queues; i++) + dev->data->rx_queue_state[i] = RTE_ETH_QUEUE_STATE_STOPPED; + + for (i = 0; i < dev->data->nb_tx_queues; i++) + dev->data->tx_queue_state[i] = RTE_ETH_QUEUE_STATE_STOPPED; + dev->data->dev_link.link_status = ETH_LINK_DOWN; + + return 0; } static int eth_dev_configure(struct rte_eth_dev *dev __rte_unused) { + int ret; + + ret = rte_mbuf_dyn_rx_timestamp_register(×tamp_dynfield_offset, + ×tamp_rx_dynflag); + if (ret != 0) { + PMD_LOG(ERR, "Failed to register Rx timestamp field/flag"); + return -rte_errno; + } + 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->driver_name = drivername; + dev_info->if_index = internals->if_index; dev_info->max_mac_addrs = 1; dev_info->max_rx_pktlen = (uint32_t) -1; dev_info->max_rx_queues = dev->data->nb_rx_queues; dev_info->max_tx_queues = dev->data->nb_tx_queues; dev_info->min_rx_bufsize = 0; - dev_info->pci_dev = NULL; + + return 0; } -static void -eth_stats_get(struct rte_eth_dev *dev, - struct rte_eth_stats *igb_stats) +static int +eth_stats_get(struct rte_eth_dev *dev, struct rte_eth_stats *stats) { - unsigned i; + unsigned int i; unsigned long rx_packets_total = 0, rx_bytes_total = 0; unsigned long tx_packets_total = 0, tx_bytes_total = 0; unsigned long tx_packets_err_total = 0; @@ -518,48 +700,90 @@ eth_stats_get(struct rte_eth_dev *dev, for (i = 0; i < RTE_ETHDEV_QUEUE_STAT_CNTRS && i < dev->data->nb_rx_queues; i++) { - igb_stats->q_ipackets[i] = internal->rx_queue[i].rx_pkts; - igb_stats->q_ibytes[i] = internal->rx_queue[i].rx_bytes; - rx_packets_total += igb_stats->q_ipackets[i]; - rx_bytes_total += igb_stats->q_ibytes[i]; + stats->q_ipackets[i] = internal->rx_queue[i].rx_stat.pkts; + stats->q_ibytes[i] = internal->rx_queue[i].rx_stat.bytes; + rx_packets_total += stats->q_ipackets[i]; + rx_bytes_total += stats->q_ibytes[i]; } for (i = 0; i < RTE_ETHDEV_QUEUE_STAT_CNTRS && i < dev->data->nb_tx_queues; i++) { - igb_stats->q_opackets[i] = internal->tx_queue[i].tx_pkts; - igb_stats->q_obytes[i] = internal->tx_queue[i].tx_bytes; - igb_stats->q_errors[i] = internal->tx_queue[i].err_pkts; - tx_packets_total += igb_stats->q_opackets[i]; - tx_bytes_total += igb_stats->q_obytes[i]; - tx_packets_err_total += igb_stats->q_errors[i]; + stats->q_opackets[i] = internal->tx_queue[i].tx_stat.pkts; + stats->q_obytes[i] = internal->tx_queue[i].tx_stat.bytes; + tx_packets_total += stats->q_opackets[i]; + tx_bytes_total += stats->q_obytes[i]; + tx_packets_err_total += internal->tx_queue[i].tx_stat.err_pkts; } - igb_stats->ipackets = rx_packets_total; - igb_stats->ibytes = rx_bytes_total; - igb_stats->opackets = tx_packets_total; - igb_stats->obytes = tx_bytes_total; - igb_stats->oerrors = tx_packets_err_total; + stats->ipackets = rx_packets_total; + stats->ibytes = rx_bytes_total; + stats->opackets = tx_packets_total; + stats->obytes = tx_bytes_total; + stats->oerrors = tx_packets_err_total; + + return 0; } -static void +static int eth_stats_reset(struct rte_eth_dev *dev) { - unsigned i; + unsigned int i; struct pmd_internals *internal = dev->data->dev_private; + for (i = 0; i < dev->data->nb_rx_queues; i++) { - internal->rx_queue[i].rx_pkts = 0; - internal->rx_queue[i].rx_bytes = 0; + internal->rx_queue[i].rx_stat.pkts = 0; + internal->rx_queue[i].rx_stat.bytes = 0; } + for (i = 0; i < dev->data->nb_tx_queues; i++) { - internal->tx_queue[i].tx_pkts = 0; - internal->tx_queue[i].tx_bytes = 0; - internal->tx_queue[i].err_pkts = 0; + internal->tx_queue[i].tx_stat.pkts = 0; + internal->tx_queue[i].tx_stat.bytes = 0; + internal->tx_queue[i].tx_stat.err_pkts = 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) { + unsigned int i; + struct pmd_internals *internals = dev->data->dev_private; + + PMD_LOG(INFO, "Closing pcap ethdev on NUMA socket %d", + rte_socket_id()); + + rte_free(dev->process_private); + + if (rte_eal_process_type() != RTE_PROC_PRIMARY) + return 0; + + /* Device wide flag, but cleanup must be performed per queue. */ + if (internals->infinite_rx) { + for (i = 0; i < dev->data->nb_rx_queues; i++) { + struct pcap_rx_queue *pcap_q = &internals->rx_queue[i]; + struct rte_mbuf *pcap_buf; + + /* + * 'pcap_q->pkts' can be NULL if 'eth_dev_close()' + * called before 'eth_rx_queue_setup()' has been called + */ + if (pcap_q->pkts == NULL) + continue; + + while (!rte_ring_dequeue(pcap_q->pkts, + (void **)&pcap_buf)) + rte_pktmbuf_free(pcap_buf); + + rte_ring_free(pcap_q->pkts); + } + } + + if (internals->phy_mac == 0) + /* not dynamically allocated, must not be freed */ + dev->data->mac_addrs = NULL; + + return 0; } static void @@ -584,9 +808,65 @@ eth_rx_queue_setup(struct rte_eth_dev *dev, { struct pmd_internals *internals = dev->data->dev_private; struct pcap_rx_queue *pcap_q = &internals->rx_queue[rx_queue_id]; + pcap_q->mb_pool = mb_pool; + pcap_q->port_id = dev->data->port_id; + pcap_q->queue_id = rx_queue_id; dev->data->rx_queues[rx_queue_id] = pcap_q; - pcap_q->in_port = dev->data->port_id; + + if (internals->infinite_rx) { + struct pmd_process_private *pp; + char ring_name[NAME_MAX]; + static uint32_t ring_number; + uint64_t pcap_pkt_count = 0; + struct rte_mbuf *bufs[1]; + pcap_t **pcap; + + pp = rte_eth_devices[pcap_q->port_id].process_private; + pcap = &pp->rx_pcap[pcap_q->queue_id]; + + if (unlikely(*pcap == NULL)) + return -ENOENT; + + pcap_pkt_count = count_packets_in_pcap(pcap, pcap_q); + + snprintf(ring_name, sizeof(ring_name), "PCAP_RING%" PRIu16, + ring_number); + + pcap_q->pkts = rte_ring_create(ring_name, + rte_align64pow2(pcap_pkt_count + 1), 0, + RING_F_SP_ENQ | RING_F_SC_DEQ); + ring_number++; + if (!pcap_q->pkts) + return -ENOENT; + + /* Fill ring with packets from PCAP file one by one. */ + while (eth_pcap_rx(pcap_q, bufs, 1)) { + /* Check for multiseg mbufs. */ + if (bufs[0]->nb_segs != 1) { + rte_pktmbuf_free(*bufs); + + while (!rte_ring_dequeue(pcap_q->pkts, + (void **)bufs)) + rte_pktmbuf_free(*bufs); + + rte_ring_free(pcap_q->pkts); + PMD_LOG(ERR, "Multiseg mbufs are not supported in infinite_rx " + "mode."); + return -EINVAL; + } + + rte_ring_enqueue_bulk(pcap_q->pkts, + (void * const *)bufs, 1, NULL); + } + /* + * Reset the stats for this queue since eth_pcap_rx calls above + * didn't result in the application receiving packets. + */ + pcap_q->rx_stat.pkts = 0; + pcap_q->rx_stat.bytes = 0; + } + return 0; } @@ -597,20 +877,60 @@ eth_tx_queue_setup(struct rte_eth_dev *dev, unsigned int socket_id __rte_unused, const struct rte_eth_txconf *tx_conf __rte_unused) { - struct pmd_internals *internals = dev->data->dev_private; - dev->data->tx_queues[tx_queue_id] = &internals->tx_queue[tx_queue_id]; + struct pcap_tx_queue *pcap_q = &internals->tx_queue[tx_queue_id]; + + pcap_q->port_id = dev->data->port_id; + pcap_q->queue_id = tx_queue_id; + dev->data->tx_queues[tx_queue_id] = pcap_q; + + return 0; +} + +static int +eth_rx_queue_start(struct rte_eth_dev *dev, uint16_t rx_queue_id) +{ + dev->data->rx_queue_state[rx_queue_id] = RTE_ETH_QUEUE_STATE_STARTED; + + return 0; +} + +static int +eth_tx_queue_start(struct rte_eth_dev *dev, uint16_t tx_queue_id) +{ + dev->data->tx_queue_state[tx_queue_id] = RTE_ETH_QUEUE_STATE_STARTED; + + return 0; +} + +static int +eth_rx_queue_stop(struct rte_eth_dev *dev, uint16_t rx_queue_id) +{ + dev->data->rx_queue_state[rx_queue_id] = RTE_ETH_QUEUE_STATE_STOPPED; + + return 0; +} + +static int +eth_tx_queue_stop(struct rte_eth_dev *dev, uint16_t tx_queue_id) +{ + dev->data->tx_queue_state[tx_queue_id] = RTE_ETH_QUEUE_STATE_STOPPED; + return 0; } static const struct eth_dev_ops ops = { .dev_start = eth_dev_start, - .dev_stop = eth_dev_stop, + .dev_stop = eth_dev_stop, .dev_close = eth_dev_close, .dev_configure = eth_dev_configure, .dev_infos_get = eth_dev_info, .rx_queue_setup = eth_rx_queue_setup, .tx_queue_setup = eth_tx_queue_setup, + .rx_queue_start = eth_rx_queue_start, + .tx_queue_start = eth_tx_queue_start, + .rx_queue_stop = eth_rx_queue_stop, + .tx_queue_stop = eth_tx_queue_stop, .rx_queue_release = eth_queue_release, .tx_queue_release = eth_queue_release, .link_update = eth_link_update, @@ -618,6 +938,22 @@ static const struct eth_dev_ops ops = { .stats_reset = eth_stats_reset, }; +static int +add_queue(struct pmd_devargs *pmd, const char *name, const char *type, + pcap_t *pcap, pcap_dumper_t *dumper) +{ + if (pmd->num_of_queue >= RTE_PMD_PCAP_MAX_QUEUES) + return -1; + if (pcap) + pmd->queue[pmd->num_of_queue].pcap = pcap; + if (dumper) + pmd->queue[pmd->num_of_queue].dumper = dumper; + pmd->queue[pmd->num_of_queue].name = name; + pmd->queue[pmd->num_of_queue].type = type; + pmd->num_of_queue++; + return 0; +} + /* * Function handler that opens the pcap file for reading a stores a * reference of it for use it later on. @@ -625,30 +961,18 @@ static const struct eth_dev_ops ops = { static int open_rx_pcap(const char *key, const char *value, void *extra_args) { - unsigned i; const char *pcap_filename = value; - struct rx_pcaps *pcaps = extra_args; + struct pmd_devargs *rx = extra_args; pcap_t *pcap = NULL; - for (i = 0; i < pcaps->num_of_rx; i++) { - if (open_single_rx_pcap(pcap_filename, &pcap) < 0) - return -1; - - pcaps->pcaps[i] = pcap; - pcaps->names[i] = pcap_filename; - pcaps->types[i] = key; - } - - return 0; -} + if (open_single_rx_pcap(pcap_filename, &pcap) < 0) + return -1; -static int -open_single_rx_pcap(const char *pcap_filename, pcap_t **pcap) -{ - if ((*pcap = pcap_open_offline(pcap_filename, errbuf)) == NULL) { - RTE_LOG(ERR, PMD, "Couldn't open %s: %s\n", pcap_filename, errbuf); + if (add_queue(rx, pcap_filename, key, pcap, NULL) < 0) { + pcap_close(pcap); return -1; } + return 0; } @@ -659,79 +983,69 @@ open_single_rx_pcap(const char *pcap_filename, pcap_t **pcap) static int open_tx_pcap(const char *key, const char *value, void *extra_args) { - unsigned i; const char *pcap_filename = value; - struct tx_pcaps *dumpers = extra_args; + struct pmd_devargs *dumpers = extra_args; pcap_dumper_t *dumper; - for (i = 0; i < dumpers->num_of_tx; i++) { - if (open_single_tx_pcap(pcap_filename, &dumper) < 0) - return -1; + if (open_single_tx_pcap(pcap_filename, &dumper) < 0) + return -1; - dumpers->dumpers[i] = dumper; - dumpers->names[i] = pcap_filename; - dumpers->types[i] = key; + if (add_queue(dumpers, pcap_filename, key, NULL, dumper) < 0) { + pcap_dump_close(dumper); + return -1; } return 0; } -static int -open_single_tx_pcap(const char *pcap_filename, pcap_dumper_t **dumper) +/* + * Opens an interface for reading and writing + */ +static inline int +open_rx_tx_iface(const char *key, const char *value, void *extra_args) { - pcap_t *tx_pcap; - /* - * We need to create a dummy empty pcap_t to use it - * with pcap_dump_open(). We create big enough an Ethernet - * pcap holder. - */ + const char *iface = value; + struct pmd_devargs *tx = extra_args; + pcap_t *pcap = NULL; - if ((tx_pcap = pcap_open_dead(DLT_EN10MB, RTE_ETH_PCAP_SNAPSHOT_LEN)) - == NULL) { - RTE_LOG(ERR, PMD, "Couldn't create dead pcap\n"); + if (open_single_iface(iface, &pcap) < 0) return -1; - } - /* The dumper is created using the previous pcap_t reference */ - if ((*dumper = pcap_dump_open(tx_pcap, pcap_filename)) == NULL) { - RTE_LOG(ERR, PMD, "Couldn't open %s for writing.\n", pcap_filename); - return -1; - } + tx->queue[0].pcap = pcap; + tx->queue[0].name = iface; + tx->queue[0].type = key; return 0; } -/* - * pcap_open_live wrapper function - */ static inline int -open_iface_live(const char *iface, pcap_t **pcap) { - *pcap = pcap_open_live(iface, RTE_ETH_PCAP_SNAPLEN, - RTE_ETH_PCAP_PROMISC, RTE_ETH_PCAP_TIMEOUT, errbuf); - - if (*pcap == NULL) { - RTE_LOG(ERR, PMD, "Couldn't open %s: %s\n", iface, errbuf); +set_iface_direction(const char *iface, pcap_t *pcap, + pcap_direction_t direction) +{ + const char *direction_str = (direction == PCAP_D_IN) ? "IN" : "OUT"; + if (pcap_setdirection(pcap, direction) < 0) { + PMD_LOG(ERR, "Setting %s pcap direction %s failed - %s\n", + iface, direction_str, pcap_geterr(pcap)); return -1; } + PMD_LOG(INFO, "Setting %s pcap direction %s\n", + iface, direction_str); return 0; } -/* - * Opens an interface for reading and writing - */ static inline int -open_rx_tx_iface(const char *key, const char *value, void *extra_args) +open_iface(const char *key, const char *value, void *extra_args) { const char *iface = value; - struct rx_pcaps *pcaps = extra_args; + struct pmd_devargs *pmd = extra_args; pcap_t *pcap = NULL; if (open_single_iface(iface, &pcap) < 0) return -1; - - pcaps->pcaps[0] = pcap; - pcaps->names[0] = iface; - pcaps->types[0] = key; + if (add_queue(pmd, iface, key, pcap, NULL) < 0) { + pcap_close(pcap); + return -1; + } return 0; } @@ -742,258 +1056,355 @@ open_rx_tx_iface(const char *key, const char *value, void *extra_args) static inline int open_rx_iface(const char *key, const char *value, void *extra_args) { - unsigned i; - const char *iface = value; - struct rx_pcaps *pcaps = extra_args; - pcap_t *pcap = NULL; + int ret = open_iface(key, value, extra_args); + if (ret < 0) + return ret; + if (strcmp(key, ETH_PCAP_RX_IFACE_IN_ARG) == 0) { + struct pmd_devargs *pmd = extra_args; + unsigned int qid = pmd->num_of_queue - 1; - for (i = 0; i < pcaps->num_of_rx; i++) { - if (open_single_iface(iface, &pcap) < 0) - return -1; - pcaps->pcaps[i] = pcap; - pcaps->names[i] = iface; - pcaps->types[i] = key; + set_iface_direction(pmd->queue[qid].name, + pmd->queue[qid].pcap, + PCAP_D_IN); } return 0; } +static inline int +rx_iface_args_process(const char *key, const char *value, void *extra_args) +{ + if (strcmp(key, ETH_PCAP_RX_IFACE_ARG) == 0 || + strcmp(key, ETH_PCAP_RX_IFACE_IN_ARG) == 0) + return open_rx_iface(key, value, extra_args); + + return 0; +} + /* * Opens a NIC for writing packets to it */ static int open_tx_iface(const char *key, const char *value, void *extra_args) { - unsigned i; - const char *iface = value; - struct tx_pcaps *pcaps = extra_args; - pcap_t *pcap; + return open_iface(key, value, extra_args); +} - for (i = 0; i < pcaps->num_of_tx; i++) { - if (open_single_iface(iface, &pcap) < 0) - return -1; - pcaps->pcaps[i] = pcap; - pcaps->names[i] = iface; - pcaps->types[i] = key; - } +static int +select_phy_mac(const char *key __rte_unused, const char *value, + void *extra_args) +{ + if (extra_args) { + const int phy_mac = atoi(value); + int *enable_phy_mac = extra_args; + if (phy_mac) + *enable_phy_mac = 1; + } return 0; } static int -open_single_iface(const char *iface, pcap_t **pcap) +get_infinite_rx_arg(const char *key __rte_unused, + const char *value, void *extra_args) { - if (open_iface_live(iface, pcap) < 0) { - RTE_LOG(ERR, PMD, "Couldn't open interface %s\n", iface); - return -1; - } + if (extra_args) { + const int infinite_rx = atoi(value); + int *enable_infinite_rx = extra_args; + if (infinite_rx > 0) + *enable_infinite_rx = 1; + } return 0; } static int -rte_pmd_init_internals(const char *name, const unsigned nb_rx_queues, - const unsigned nb_tx_queues, - const unsigned numa_node, +pmd_init_internals(struct rte_vdev_device *vdev, + const unsigned int nb_rx_queues, + const unsigned int nb_tx_queues, struct pmd_internals **internals, - struct rte_eth_dev **eth_dev, - struct rte_kvargs *kvlist) + struct rte_eth_dev **eth_dev) { - struct rte_eth_dev_data *data = NULL; - unsigned k_idx; - struct rte_kvargs_pair *pair = NULL; - - for (k_idx = 0; k_idx < kvlist->count; k_idx++) { - pair = &kvlist->pairs[k_idx]; - if (strstr(pair->key, ETH_PCAP_IFACE_ARG) != NULL) - break; - } + struct rte_eth_dev_data *data; + struct pmd_process_private *pp; + unsigned int numa_node = vdev->device.numa_node; - RTE_LOG(INFO, PMD, - "Creating pcap-backed ethdev on numa socket %u\n", numa_node); + PMD_LOG(INFO, "Creating pcap-backed ethdev on numa socket %d", + numa_node); - /* now do all data allocation - for eth_dev structure - * and internal (private) data - */ - data = rte_zmalloc_socket(name, sizeof(*data), 0, numa_node); - if (data == NULL) - goto error; + pp = (struct pmd_process_private *) + rte_zmalloc(NULL, sizeof(struct pmd_process_private), + RTE_CACHE_LINE_SIZE); - *internals = rte_zmalloc_socket(name, sizeof(**internals), 0, numa_node); - if (*internals == NULL) - goto error; + if (pp == NULL) { + PMD_LOG(ERR, + "Failed to allocate memory for process private"); + return -1; + } /* reserve an ethdev entry */ - *eth_dev = rte_eth_dev_allocate(name); - if (*eth_dev == NULL) - goto error; - - /* check length of device name */ - if ((strlen((*eth_dev)->data->name) + 1) > sizeof(data->name)) - goto error; - + *eth_dev = rte_eth_vdev_allocate(vdev, sizeof(**internals)); + if (!(*eth_dev)) { + rte_free(pp); + return -1; + } + (*eth_dev)->process_private = pp; /* now put it all together * - store queue data in internals, * - store numa_node info in eth_dev * - point eth_dev_data to internals * - and point eth_dev structure to new eth_dev_data structure */ - /* NOTE: we'll replace the data element, of originally allocated eth_dev - * so the rings are local per-process */ - - if (pair == NULL) - (*internals)->if_index = 0; - else - (*internals)->if_index = if_nametoindex(pair->value); - - data->dev_private = *internals; - data->port_id = (*eth_dev)->data->port_id; - snprintf(data->name, sizeof(data->name), "%s", (*eth_dev)->data->name); + *internals = (*eth_dev)->data->dev_private; + /* + * Interface MAC = 02:70:63:61:70: + * derived from: 'locally administered':'p':'c':'a':'p':'iface_idx' + * where the middle 4 characters are converted to hex. + */ + (*internals)->eth_addr = (struct rte_ether_addr) { + .addr_bytes = { 0x02, 0x70, 0x63, 0x61, 0x70, iface_idx++ } + }; + (*internals)->phy_mac = 0; + data = (*eth_dev)->data; data->nb_rx_queues = (uint16_t)nb_rx_queues; data->nb_tx_queues = (uint16_t)nb_tx_queues; data->dev_link = pmd_link; - data->mac_addrs = ð_addr; - strncpy(data->name, - (*eth_dev)->data->name, strlen((*eth_dev)->data->name)); + data->mac_addrs = &(*internals)->eth_addr; + data->promiscuous = 1; + data->all_multicast = 1; + data->dev_flags |= RTE_ETH_DEV_AUTOFILL_QUEUE_XSTATS; - (*eth_dev)->data = data; + /* + * NOTE: we'll replace the data element, of originally allocated + * eth_dev so the rings are local per-process + */ (*eth_dev)->dev_ops = &ops; - (*eth_dev)->driver = NULL; - data->dev_flags = RTE_ETH_DEV_DETACHABLE; - data->kdrv = RTE_KDRV_NONE; - data->drv_name = drivername; - data->numa_node = numa_node; - return 0; - -error: - rte_free(data); - rte_free(*internals); + strlcpy((*internals)->devargs, rte_vdev_device_args(vdev), + ETH_PCAP_ARG_MAXLEN); - return -1; + return 0; } static int -rte_eth_from_pcaps_common(const char *name, struct rx_pcaps *rx_queues, - const unsigned nb_rx_queues, struct tx_pcaps *tx_queues, - const unsigned nb_tx_queues, const unsigned numa_node, - struct rte_kvargs *kvlist, struct pmd_internals **internals, - struct rte_eth_dev **eth_dev) +eth_pcap_update_mac(const char *if_name, struct rte_eth_dev *eth_dev, + const unsigned int numa_node) { - unsigned i; +#if defined(RTE_EXEC_ENV_LINUX) + void *mac_addrs; + struct ifreq ifr; + int if_fd = socket(AF_INET, SOCK_DGRAM, 0); + + if (if_fd == -1) + return -1; - /* do some parameter checking */ - if (rx_queues == NULL && nb_rx_queues > 0) + rte_strscpy(ifr.ifr_name, if_name, sizeof(ifr.ifr_name)); + if (ioctl(if_fd, SIOCGIFHWADDR, &ifr)) { + close(if_fd); return -1; - if (tx_queues == NULL && nb_tx_queues > 0) + } + + mac_addrs = rte_zmalloc_socket(NULL, RTE_ETHER_ADDR_LEN, 0, numa_node); + if (!mac_addrs) { + close(if_fd); + return -1; + } + + PMD_LOG(INFO, "Setting phy MAC for %s", if_name); + eth_dev->data->mac_addrs = mac_addrs; + rte_memcpy(eth_dev->data->mac_addrs[0].addr_bytes, + ifr.ifr_hwaddr.sa_data, RTE_ETHER_ADDR_LEN); + + close(if_fd); + + return 0; + +#elif defined(RTE_EXEC_ENV_FREEBSD) + void *mac_addrs; + struct if_msghdr *ifm; + struct sockaddr_dl *sdl; + int mib[6]; + size_t len = 0; + char *buf; + + mib[0] = CTL_NET; + mib[1] = AF_ROUTE; + mib[2] = 0; + mib[3] = AF_LINK; + mib[4] = NET_RT_IFLIST; + mib[5] = if_nametoindex(if_name); + + if (sysctl(mib, 6, NULL, &len, NULL, 0) < 0) return -1; - if (rte_pmd_init_internals(name, nb_rx_queues, nb_tx_queues, numa_node, - internals, eth_dev, kvlist) < 0) + if (len == 0) return -1; - for (i = 0; i < nb_rx_queues; i++) { - (*internals)->rx_queue[i].pcap = rx_queues->pcaps[i]; - snprintf((*internals)->rx_queue[i].name, - sizeof((*internals)->rx_queue[i].name), "%s", - rx_queues->names[i]); - snprintf((*internals)->rx_queue[i].type, - sizeof((*internals)->rx_queue[i].type), "%s", - rx_queues->types[i]); + buf = rte_malloc(NULL, len, 0); + if (!buf) + return -1; + + if (sysctl(mib, 6, buf, &len, NULL, 0) < 0) { + rte_free(buf); + return -1; } - for (i = 0; i < nb_tx_queues; i++) { - (*internals)->tx_queue[i].dumper = tx_queues->dumpers[i]; - snprintf((*internals)->tx_queue[i].name, - sizeof((*internals)->tx_queue[i].name), "%s", - tx_queues->names[i]); - snprintf((*internals)->tx_queue[i].type, - sizeof((*internals)->tx_queue[i].type), "%s", - tx_queues->types[i]); + ifm = (struct if_msghdr *)buf; + sdl = (struct sockaddr_dl *)(ifm + 1); + + mac_addrs = rte_zmalloc_socket(NULL, RTE_ETHER_ADDR_LEN, 0, numa_node); + if (!mac_addrs) { + rte_free(buf); + return -1; } + PMD_LOG(INFO, "Setting phy MAC for %s", if_name); + eth_dev->data->mac_addrs = mac_addrs; + rte_memcpy(eth_dev->data->mac_addrs[0].addr_bytes, + LLADDR(sdl), RTE_ETHER_ADDR_LEN); + + rte_free(buf); + return 0; +#else + return -1; +#endif } static int -rte_eth_from_pcaps_n_dumpers(const char *name, - struct rx_pcaps *rx_queues, - const unsigned nb_rx_queues, - struct tx_pcaps *tx_queues, - const unsigned nb_tx_queues, - const unsigned numa_node, - struct rte_kvargs *kvlist) +eth_from_pcaps_common(struct rte_vdev_device *vdev, + struct pmd_devargs_all *devargs_all, + struct pmd_internals **internals, struct rte_eth_dev **eth_dev) { - struct pmd_internals *internals = NULL; - struct rte_eth_dev *eth_dev = NULL; - int ret; + struct pmd_process_private *pp; + struct pmd_devargs *rx_queues = &devargs_all->rx_queues; + struct pmd_devargs *tx_queues = &devargs_all->tx_queues; + const unsigned int nb_rx_queues = rx_queues->num_of_queue; + const unsigned int nb_tx_queues = tx_queues->num_of_queue; + unsigned int i; + + if (pmd_init_internals(vdev, nb_rx_queues, nb_tx_queues, internals, + eth_dev) < 0) + return -1; - ret = rte_eth_from_pcaps_common(name, rx_queues, nb_rx_queues, - tx_queues, nb_tx_queues, numa_node, kvlist, - &internals, ð_dev); + pp = (*eth_dev)->process_private; + for (i = 0; i < nb_rx_queues; i++) { + struct pcap_rx_queue *rx = &(*internals)->rx_queue[i]; + struct devargs_queue *queue = &rx_queues->queue[i]; - if (ret < 0) - return ret; + pp->rx_pcap[i] = queue->pcap; + strlcpy(rx->name, queue->name, sizeof(rx->name)); + strlcpy(rx->type, queue->type, sizeof(rx->type)); + } - /* using multiple pcaps/interfaces */ - internals->single_iface = 0; + for (i = 0; i < nb_tx_queues; i++) { + struct pcap_tx_queue *tx = &(*internals)->tx_queue[i]; + struct devargs_queue *queue = &tx_queues->queue[i]; - eth_dev->rx_pkt_burst = eth_pcap_rx; - eth_dev->tx_pkt_burst = eth_pcap_tx_dumper; + pp->tx_dumper[i] = queue->dumper; + pp->tx_pcap[i] = queue->pcap; + strlcpy(tx->name, queue->name, sizeof(tx->name)); + strlcpy(tx->type, queue->type, sizeof(tx->type)); + } return 0; } static int -rte_eth_from_pcaps(const char *name, - struct rx_pcaps *rx_queues, - const unsigned nb_rx_queues, - struct tx_pcaps *tx_queues, - const unsigned nb_tx_queues, - const unsigned numa_node, - struct rte_kvargs *kvlist, - int single_iface) +eth_from_pcaps(struct rte_vdev_device *vdev, + struct pmd_devargs_all *devargs_all) { struct pmd_internals *internals = NULL; struct rte_eth_dev *eth_dev = NULL; + struct pmd_devargs *rx_queues = &devargs_all->rx_queues; + int single_iface = devargs_all->single_iface; + unsigned int infinite_rx = devargs_all->infinite_rx; int ret; - ret = rte_eth_from_pcaps_common(name, rx_queues, nb_rx_queues, - tx_queues, nb_tx_queues, numa_node, kvlist, - &internals, ð_dev); + ret = eth_from_pcaps_common(vdev, devargs_all, &internals, ð_dev); if (ret < 0) return ret; - /* store wether we are using a single interface for rx/tx or not */ + /* store weather we are using a single interface for rx/tx or not */ internals->single_iface = single_iface; - eth_dev->rx_pkt_burst = eth_pcap_rx; - eth_dev->tx_pkt_burst = eth_pcap_tx; + if (single_iface) { + internals->if_index = if_nametoindex(rx_queues->queue[0].name); + + /* phy_mac arg is applied only only if "iface" devarg is provided */ + if (rx_queues->phy_mac) { + int ret = eth_pcap_update_mac(rx_queues->queue[0].name, + eth_dev, vdev->device.numa_node); + if (ret == 0) + internals->phy_mac = 1; + } + } + + internals->infinite_rx = infinite_rx; + /* Assign rx ops. */ + if (infinite_rx) + eth_dev->rx_pkt_burst = eth_pcap_rx_infinite; + else if (devargs_all->is_rx_pcap || devargs_all->is_rx_iface || + single_iface) + eth_dev->rx_pkt_burst = eth_pcap_rx; + else + eth_dev->rx_pkt_burst = eth_null_rx; + + /* Assign tx ops. */ + if (devargs_all->is_tx_pcap) + eth_dev->tx_pkt_burst = eth_pcap_tx_dumper; + else if (devargs_all->is_tx_iface || single_iface) + eth_dev->tx_pkt_burst = eth_pcap_tx; + else + eth_dev->tx_pkt_burst = eth_tx_drop; + rte_eth_dev_probing_finish(eth_dev); return 0; } - static int -rte_pmd_pcap_devinit(const char *name, const char *params) +pmd_pcap_probe(struct rte_vdev_device *dev) { - unsigned numa_node, using_dumpers = 0; - int ret; + const char *name; struct rte_kvargs *kvlist; - struct rx_pcaps pcaps = {0}; - struct tx_pcaps dumpers = {0}; - - RTE_LOG(INFO, PMD, "Initializing pmd_pcap for %s\n", name); - - numa_node = rte_socket_id(); + struct pmd_devargs pcaps = {0}; + struct pmd_devargs dumpers = {0}; + struct rte_eth_dev *eth_dev = NULL; + struct pmd_internals *internal; + int ret = 0; + + struct pmd_devargs_all devargs_all = { + .single_iface = 0, + .is_tx_pcap = 0, + .is_tx_iface = 0, + .infinite_rx = 0, + }; + + name = rte_vdev_device_name(dev); + PMD_LOG(INFO, "Initializing pmd_pcap for %s", name); gettimeofday(&start_time, NULL); start_cycles = rte_get_timer_cycles(); hz = rte_get_timer_hz(); - kvlist = rte_kvargs_parse(params, valid_arguments); - if (kvlist == NULL) - return -1; + 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); + return -1; + } + + internal = eth_dev->data->dev_private; + + kvlist = rte_kvargs_parse(internal->devargs, valid_arguments); + if (kvlist == NULL) + return -1; + } else { + kvlist = rte_kvargs_parse(rte_vdev_device_args(dev), + valid_arguments); + if (kvlist == NULL) + return -1; + } /* * If iface argument is passed we open the NICs and use them for @@ -1005,95 +1416,199 @@ rte_pmd_pcap_devinit(const char *name, const char *params) &open_rx_tx_iface, &pcaps); if (ret < 0) goto free_kvlist; - dumpers.pcaps[0] = pcaps.pcaps[0]; - dumpers.names[0] = pcaps.names[0]; - dumpers.types[0] = pcaps.types[0]; - ret = rte_eth_from_pcaps(name, &pcaps, 1, &dumpers, 1, - numa_node, kvlist, 1); - goto free_kvlist; + + dumpers.queue[0] = pcaps.queue[0]; + + ret = rte_kvargs_process(kvlist, ETH_PCAP_PHY_MAC_ARG, + &select_phy_mac, &pcaps.phy_mac); + if (ret < 0) + goto free_kvlist; + + dumpers.phy_mac = pcaps.phy_mac; + + devargs_all.single_iface = 1; + pcaps.num_of_queue = 1; + dumpers.num_of_queue = 1; + + goto create_eth; } /* - * We check whether we want to open a RX stream from a real NIC or a - * pcap file + * We check whether we want to open a RX stream from a real NIC, a + * pcap file or open a dummy RX stream */ - if ((pcaps.num_of_rx = rte_kvargs_count(kvlist, ETH_PCAP_RX_PCAP_ARG))) { + devargs_all.is_rx_pcap = + rte_kvargs_count(kvlist, ETH_PCAP_RX_PCAP_ARG) ? 1 : 0; + devargs_all.is_rx_iface = + (rte_kvargs_count(kvlist, ETH_PCAP_RX_IFACE_ARG) + + rte_kvargs_count(kvlist, ETH_PCAP_RX_IFACE_IN_ARG)) ? 1 : 0; + pcaps.num_of_queue = 0; + + devargs_all.is_tx_pcap = + rte_kvargs_count(kvlist, ETH_PCAP_TX_PCAP_ARG) ? 1 : 0; + devargs_all.is_tx_iface = + rte_kvargs_count(kvlist, ETH_PCAP_TX_IFACE_ARG) ? 1 : 0; + dumpers.num_of_queue = 0; + + if (devargs_all.is_rx_pcap) { + /* + * We check whether we want to infinitely rx the pcap file. + */ + unsigned int infinite_rx_arg_cnt = rte_kvargs_count(kvlist, + ETH_PCAP_INFINITE_RX_ARG); + + if (infinite_rx_arg_cnt == 1) { + ret = rte_kvargs_process(kvlist, + ETH_PCAP_INFINITE_RX_ARG, + &get_infinite_rx_arg, + &devargs_all.infinite_rx); + if (ret < 0) + goto free_kvlist; + PMD_LOG(INFO, "infinite_rx has been %s for %s", + devargs_all.infinite_rx ? "enabled" : "disabled", + name); + + } else if (infinite_rx_arg_cnt > 1) { + PMD_LOG(WARNING, "infinite_rx has not been enabled since the " + "argument has been provided more than once " + "for %s", name); + } + ret = rte_kvargs_process(kvlist, ETH_PCAP_RX_PCAP_ARG, &open_rx_pcap, &pcaps); + } else if (devargs_all.is_rx_iface) { + ret = rte_kvargs_process(kvlist, NULL, + &rx_iface_args_process, &pcaps); + } else if (devargs_all.is_tx_iface || devargs_all.is_tx_pcap) { + unsigned int i; + + /* Count number of tx queue args passed before dummy rx queue + * creation so a dummy rx queue can be created for each tx queue + */ + unsigned int num_tx_queues = + (rte_kvargs_count(kvlist, ETH_PCAP_TX_PCAP_ARG) + + rte_kvargs_count(kvlist, ETH_PCAP_TX_IFACE_ARG)); + + PMD_LOG(INFO, "Creating null rx queue since no rx queues were provided."); + + /* Creating a dummy rx queue for each tx queue passed */ + for (i = 0; i < num_tx_queues; i++) + ret = add_queue(&pcaps, "dummy_rx", "rx_null", NULL, + NULL); } else { - pcaps.num_of_rx = rte_kvargs_count(kvlist, - ETH_PCAP_RX_IFACE_ARG); - ret = rte_kvargs_process(kvlist, ETH_PCAP_RX_IFACE_ARG, - &open_rx_iface, &pcaps); + PMD_LOG(ERR, "Error - No rx or tx queues provided"); + ret = -ENOENT; } - if (ret < 0) goto free_kvlist; /* - * We check whether we want to open a TX stream to a real NIC or a - * pcap file + * We check whether we want to open a TX stream to a real NIC, + * a pcap file, or drop packets on tx */ - if ((dumpers.num_of_tx = rte_kvargs_count(kvlist, - ETH_PCAP_TX_PCAP_ARG))) { + if (devargs_all.is_tx_pcap) { ret = rte_kvargs_process(kvlist, ETH_PCAP_TX_PCAP_ARG, &open_tx_pcap, &dumpers); - using_dumpers = 1; - } else { - dumpers.num_of_tx = rte_kvargs_count(kvlist, - ETH_PCAP_TX_IFACE_ARG); + } else if (devargs_all.is_tx_iface) { ret = rte_kvargs_process(kvlist, ETH_PCAP_TX_IFACE_ARG, &open_tx_iface, &dumpers); + } else { + unsigned int i; + + PMD_LOG(INFO, "Dropping packets on tx since no tx queues were provided."); + + /* Add 1 dummy queue per rxq which counts and drops packets. */ + for (i = 0; i < pcaps.num_of_queue; i++) + ret = add_queue(&dumpers, "dummy_tx", "tx_drop", NULL, + NULL); } if (ret < 0) goto free_kvlist; - if (using_dumpers) - ret = rte_eth_from_pcaps_n_dumpers(name, &pcaps, pcaps.num_of_rx, - &dumpers, dumpers.num_of_tx, numa_node, kvlist); - else - ret = rte_eth_from_pcaps(name, &pcaps, pcaps.num_of_rx, &dumpers, - dumpers.num_of_tx, numa_node, kvlist, 0); +create_eth: + if (rte_eal_process_type() == RTE_PROC_SECONDARY) { + struct pmd_process_private *pp; + unsigned int i; + + internal = eth_dev->data->dev_private; + pp = (struct pmd_process_private *) + rte_zmalloc(NULL, + sizeof(struct pmd_process_private), + RTE_CACHE_LINE_SIZE); + + if (pp == NULL) { + PMD_LOG(ERR, + "Failed to allocate memory for process private"); + ret = -1; + goto free_kvlist; + } + + eth_dev->dev_ops = &ops; + eth_dev->device = &dev->device; + + /* setup process private */ + for (i = 0; i < pcaps.num_of_queue; i++) + pp->rx_pcap[i] = pcaps.queue[i].pcap; + + for (i = 0; i < dumpers.num_of_queue; i++) { + pp->tx_dumper[i] = dumpers.queue[i].dumper; + pp->tx_pcap[i] = dumpers.queue[i].pcap; + } + + eth_dev->process_private = pp; + eth_dev->rx_pkt_burst = eth_pcap_rx; + if (devargs_all.is_tx_pcap) + eth_dev->tx_pkt_burst = eth_pcap_tx_dumper; + else + eth_dev->tx_pkt_burst = eth_pcap_tx; + + rte_eth_dev_probing_finish(eth_dev); + goto free_kvlist; + } + + devargs_all.rx_queues = pcaps; + devargs_all.tx_queues = dumpers; + + ret = eth_from_pcaps(dev, &devargs_all); free_kvlist: rte_kvargs_free(kvlist); + return ret; } static int -rte_pmd_pcap_devuninit(const char *name) +pmd_pcap_remove(struct rte_vdev_device *dev) { struct rte_eth_dev *eth_dev = NULL; - RTE_LOG(INFO, PMD, "Closing pcap ethdev on numa socket %u\n", - rte_socket_id()); - - if (name == NULL) + if (!dev) return -1; - /* reserve an ethdev entry */ - eth_dev = rte_eth_dev_allocated(name); + eth_dev = rte_eth_dev_allocated(rte_vdev_device_name(dev)); if (eth_dev == NULL) - return -1; - - rte_free(eth_dev->data->dev_private); - rte_free(eth_dev->data); + return 0; /* port already released */ + eth_dev_close(eth_dev); rte_eth_dev_release_port(eth_dev); return 0; } static struct rte_vdev_driver pmd_pcap_drv = { - .init = rte_pmd_pcap_devinit, - .uninit = rte_pmd_pcap_devuninit, + .probe = pmd_pcap_probe, + .remove = pmd_pcap_remove, }; -DRIVER_REGISTER_VDEV(net_pcap, pmd_pcap_drv); -DRIVER_REGISTER_PARAM_STRING(net_pcap, - "rx_pcap= " - "tx_pcap= " - "rx_iface= " - "tx_iface= " - "iface="); +RTE_PMD_REGISTER_VDEV(net_pcap, pmd_pcap_drv); +RTE_PMD_REGISTER_ALIAS(net_pcap, eth_pcap); +RTE_PMD_REGISTER_PARAM_STRING(net_pcap, + ETH_PCAP_RX_PCAP_ARG "= " + ETH_PCAP_TX_PCAP_ARG "= " + ETH_PCAP_RX_IFACE_ARG "= " + ETH_PCAP_RX_IFACE_IN_ARG "= " + ETH_PCAP_TX_IFACE_ARG "= " + ETH_PCAP_IFACE_ARG "= " + ETH_PCAP_PHY_MAC_ARG "=" + ETH_PCAP_INFINITE_RX_ARG "=<0|1>");