numam-dpdk/app/test-pmd/config.c
Helin Zhang 19b16e2f64 ethdev: add vlan type when setting ether type
In order to set ether type of VLAN for single VLAN, inner
and outer VLAN, the VLAN type as an input parameter is added
to 'rte_eth_dev_set_vlan_ether_type()'.
In addition, corresponding changes in e1000, ixgbe and i40e
are also added.

It is an ABI break but ethdev library is already bumped for 16.04.

Signed-off-by: Helin Zhang <helin.zhang@intel.com>
Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>
2016-03-11 22:21:06 +01:00

2482 lines
67 KiB
C

/*-
* 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.
*/
/* BSD LICENSE
*
* Copyright 2013-2014 6WIND S.A.
*
* 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 6WIND S.A. 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 <stdarg.h>
#include <errno.h>
#include <stdio.h>
#include <string.h>
#include <stdarg.h>
#include <stdint.h>
#include <inttypes.h>
#include <sys/queue.h>
#include <rte_common.h>
#include <rte_byteorder.h>
#include <rte_debug.h>
#include <rte_log.h>
#include <rte_memory.h>
#include <rte_memcpy.h>
#include <rte_memzone.h>
#include <rte_launch.h>
#include <rte_eal.h>
#include <rte_per_lcore.h>
#include <rte_lcore.h>
#include <rte_atomic.h>
#include <rte_branch_prediction.h>
#include <rte_ring.h>
#include <rte_mempool.h>
#include <rte_mbuf.h>
#include <rte_interrupts.h>
#include <rte_pci.h>
#include <rte_ether.h>
#include <rte_ethdev.h>
#include <rte_string_fns.h>
#include "testpmd.h"
static char *flowtype_to_str(uint16_t flow_type);
static const struct {
enum tx_pkt_split split;
const char *name;
} tx_split_name[] = {
{
.split = TX_PKT_SPLIT_OFF,
.name = "off",
},
{
.split = TX_PKT_SPLIT_ON,
.name = "on",
},
{
.split = TX_PKT_SPLIT_RND,
.name = "rand",
},
};
struct rss_type_info {
char str[32];
uint64_t rss_type;
};
static const struct rss_type_info rss_type_table[] = {
{ "ipv4", ETH_RSS_IPV4 },
{ "ipv4-frag", ETH_RSS_FRAG_IPV4 },
{ "ipv4-tcp", ETH_RSS_NONFRAG_IPV4_TCP },
{ "ipv4-udp", ETH_RSS_NONFRAG_IPV4_UDP },
{ "ipv4-sctp", ETH_RSS_NONFRAG_IPV4_SCTP },
{ "ipv4-other", ETH_RSS_NONFRAG_IPV4_OTHER },
{ "ipv6", ETH_RSS_IPV6 },
{ "ipv6-frag", ETH_RSS_FRAG_IPV6 },
{ "ipv6-tcp", ETH_RSS_NONFRAG_IPV6_TCP },
{ "ipv6-udp", ETH_RSS_NONFRAG_IPV6_UDP },
{ "ipv6-sctp", ETH_RSS_NONFRAG_IPV6_SCTP },
{ "ipv6-other", ETH_RSS_NONFRAG_IPV6_OTHER },
{ "l2-payload", ETH_RSS_L2_PAYLOAD },
{ "ipv6-ex", ETH_RSS_IPV6_EX },
{ "ipv6-tcp-ex", ETH_RSS_IPV6_TCP_EX },
{ "ipv6-udp-ex", ETH_RSS_IPV6_UDP_EX },
};
static void
print_ethaddr(const char *name, struct ether_addr *eth_addr)
{
char buf[ETHER_ADDR_FMT_SIZE];
ether_format_addr(buf, ETHER_ADDR_FMT_SIZE, eth_addr);
printf("%s%s", name, buf);
}
void
nic_stats_display(portid_t port_id)
{
struct rte_eth_stats stats;
struct rte_port *port = &ports[port_id];
uint8_t i;
portid_t pid;
static const char *nic_stats_border = "########################";
if (port_id_is_invalid(port_id, ENABLED_WARN)) {
printf("Valid port range is [0");
FOREACH_PORT(pid, ports)
printf(", %d", pid);
printf("]\n");
return;
}
rte_eth_stats_get(port_id, &stats);
printf("\n %s NIC statistics for port %-2d %s\n",
nic_stats_border, port_id, nic_stats_border);
if ((!port->rx_queue_stats_mapping_enabled) && (!port->tx_queue_stats_mapping_enabled)) {
printf(" RX-packets: %-10"PRIu64" RX-missed: %-10"PRIu64" RX-bytes: "
"%-"PRIu64"\n",
stats.ipackets, stats.imissed, stats.ibytes);
printf(" RX-errors: %-"PRIu64"\n", stats.ierrors);
printf(" RX-nombuf: %-10"PRIu64"\n",
stats.rx_nombuf);
printf(" TX-packets: %-10"PRIu64" TX-errors: %-10"PRIu64" TX-bytes: "
"%-"PRIu64"\n",
stats.opackets, stats.oerrors, stats.obytes);
}
else {
printf(" RX-packets: %10"PRIu64" RX-errors: %10"PRIu64
" RX-bytes: %10"PRIu64"\n",
stats.ipackets, stats.ierrors, stats.ibytes);
printf(" RX-errors: %10"PRIu64"\n", stats.ierrors);
printf(" RX-nombuf: %10"PRIu64"\n",
stats.rx_nombuf);
printf(" TX-packets: %10"PRIu64" TX-errors: %10"PRIu64
" TX-bytes: %10"PRIu64"\n",
stats.opackets, stats.oerrors, stats.obytes);
}
if (port->rx_queue_stats_mapping_enabled) {
printf("\n");
for (i = 0; i < RTE_ETHDEV_QUEUE_STAT_CNTRS; i++) {
printf(" Stats reg %2d RX-packets: %10"PRIu64
" RX-errors: %10"PRIu64
" RX-bytes: %10"PRIu64"\n",
i, stats.q_ipackets[i], stats.q_errors[i], stats.q_ibytes[i]);
}
}
if (port->tx_queue_stats_mapping_enabled) {
printf("\n");
for (i = 0; i < RTE_ETHDEV_QUEUE_STAT_CNTRS; i++) {
printf(" Stats reg %2d TX-packets: %10"PRIu64
" TX-bytes: %10"PRIu64"\n",
i, stats.q_opackets[i], stats.q_obytes[i]);
}
}
printf(" %s############################%s\n",
nic_stats_border, nic_stats_border);
}
void
nic_stats_clear(portid_t port_id)
{
portid_t pid;
if (port_id_is_invalid(port_id, ENABLED_WARN)) {
printf("Valid port range is [0");
FOREACH_PORT(pid, ports)
printf(", %d", pid);
printf("]\n");
return;
}
rte_eth_stats_reset(port_id);
printf("\n NIC statistics for port %d cleared\n", port_id);
}
void
nic_xstats_display(portid_t port_id)
{
struct rte_eth_xstats *xstats;
int len, ret, i;
printf("###### NIC extended statistics for port %-2d\n", port_id);
len = rte_eth_xstats_get(port_id, NULL, 0);
if (len < 0) {
printf("Cannot get xstats count\n");
return;
}
xstats = malloc(sizeof(xstats[0]) * len);
if (xstats == NULL) {
printf("Cannot allocate memory for xstats\n");
return;
}
ret = rte_eth_xstats_get(port_id, xstats, len);
if (ret < 0 || ret > len) {
printf("Cannot get xstats\n");
free(xstats);
return;
}
for (i = 0; i < len; i++)
printf("%s: %"PRIu64"\n", xstats[i].name, xstats[i].value);
free(xstats);
}
void
nic_xstats_clear(portid_t port_id)
{
rte_eth_xstats_reset(port_id);
}
void
nic_stats_mapping_display(portid_t port_id)
{
struct rte_port *port = &ports[port_id];
uint16_t i;
portid_t pid;
static const char *nic_stats_mapping_border = "########################";
if (port_id_is_invalid(port_id, ENABLED_WARN)) {
printf("Valid port range is [0");
FOREACH_PORT(pid, ports)
printf(", %d", pid);
printf("]\n");
return;
}
if ((!port->rx_queue_stats_mapping_enabled) && (!port->tx_queue_stats_mapping_enabled)) {
printf("Port id %d - either does not support queue statistic mapping or"
" no queue statistic mapping set\n", port_id);
return;
}
printf("\n %s NIC statistics mapping for port %-2d %s\n",
nic_stats_mapping_border, port_id, nic_stats_mapping_border);
if (port->rx_queue_stats_mapping_enabled) {
for (i = 0; i < nb_rx_queue_stats_mappings; i++) {
if (rx_queue_stats_mappings[i].port_id == port_id) {
printf(" RX-queue %2d mapped to Stats Reg %2d\n",
rx_queue_stats_mappings[i].queue_id,
rx_queue_stats_mappings[i].stats_counter_id);
}
}
printf("\n");
}
if (port->tx_queue_stats_mapping_enabled) {
for (i = 0; i < nb_tx_queue_stats_mappings; i++) {
if (tx_queue_stats_mappings[i].port_id == port_id) {
printf(" TX-queue %2d mapped to Stats Reg %2d\n",
tx_queue_stats_mappings[i].queue_id,
tx_queue_stats_mappings[i].stats_counter_id);
}
}
}
printf(" %s####################################%s\n",
nic_stats_mapping_border, nic_stats_mapping_border);
}
void
rx_queue_infos_display(portid_t port_id, uint16_t queue_id)
{
struct rte_eth_rxq_info qinfo;
int32_t rc;
static const char *info_border = "*********************";
rc = rte_eth_rx_queue_info_get(port_id, queue_id, &qinfo);
if (rc != 0) {
printf("Failed to retrieve information for port: %hhu, "
"RX queue: %hu\nerror desc: %s(%d)\n",
port_id, queue_id, strerror(-rc), rc);
return;
}
printf("\n%s Infos for port %-2u, RX queue %-2u %s",
info_border, port_id, queue_id, info_border);
printf("\nMempool: %s", (qinfo.mp == NULL) ? "NULL" : qinfo.mp->name);
printf("\nRX prefetch threshold: %hhu", qinfo.conf.rx_thresh.pthresh);
printf("\nRX host threshold: %hhu", qinfo.conf.rx_thresh.hthresh);
printf("\nRX writeback threshold: %hhu", qinfo.conf.rx_thresh.wthresh);
printf("\nRX free threshold: %hu", qinfo.conf.rx_free_thresh);
printf("\nRX drop packets: %s",
(qinfo.conf.rx_drop_en != 0) ? "on" : "off");
printf("\nRX deferred start: %s",
(qinfo.conf.rx_deferred_start != 0) ? "on" : "off");
printf("\nRX scattered packets: %s",
(qinfo.scattered_rx != 0) ? "on" : "off");
printf("\nNumber of RXDs: %hu", qinfo.nb_desc);
printf("\n");
}
void
tx_queue_infos_display(portid_t port_id, uint16_t queue_id)
{
struct rte_eth_txq_info qinfo;
int32_t rc;
static const char *info_border = "*********************";
rc = rte_eth_tx_queue_info_get(port_id, queue_id, &qinfo);
if (rc != 0) {
printf("Failed to retrieve information for port: %hhu, "
"TX queue: %hu\nerror desc: %s(%d)\n",
port_id, queue_id, strerror(-rc), rc);
return;
}
printf("\n%s Infos for port %-2u, TX queue %-2u %s",
info_border, port_id, queue_id, info_border);
printf("\nTX prefetch threshold: %hhu", qinfo.conf.tx_thresh.pthresh);
printf("\nTX host threshold: %hhu", qinfo.conf.tx_thresh.hthresh);
printf("\nTX writeback threshold: %hhu", qinfo.conf.tx_thresh.wthresh);
printf("\nTX RS threshold: %hu", qinfo.conf.tx_rs_thresh);
printf("\nTX free threshold: %hu", qinfo.conf.tx_free_thresh);
printf("\nTX flags: %#x", qinfo.conf.txq_flags);
printf("\nTX deferred start: %s",
(qinfo.conf.tx_deferred_start != 0) ? "on" : "off");
printf("\nNumber of TXDs: %hu", qinfo.nb_desc);
printf("\n");
}
void
port_infos_display(portid_t port_id)
{
struct rte_port *port;
struct ether_addr mac_addr;
struct rte_eth_link link;
struct rte_eth_dev_info dev_info;
int vlan_offload;
struct rte_mempool * mp;
static const char *info_border = "*********************";
portid_t pid;
if (port_id_is_invalid(port_id, ENABLED_WARN)) {
printf("Valid port range is [0");
FOREACH_PORT(pid, ports)
printf(", %d", pid);
printf("]\n");
return;
}
port = &ports[port_id];
rte_eth_link_get_nowait(port_id, &link);
printf("\n%s Infos for port %-2d %s\n",
info_border, port_id, info_border);
rte_eth_macaddr_get(port_id, &mac_addr);
print_ethaddr("MAC address: ", &mac_addr);
printf("\nConnect to socket: %u", port->socket_id);
if (port_numa[port_id] != NUMA_NO_CONFIG) {
mp = mbuf_pool_find(port_numa[port_id]);
if (mp)
printf("\nmemory allocation on the socket: %d",
port_numa[port_id]);
} else
printf("\nmemory allocation on the socket: %u",port->socket_id);
printf("\nLink status: %s\n", (link.link_status) ? ("up") : ("down"));
printf("Link speed: %u Mbps\n", (unsigned) link.link_speed);
printf("Link duplex: %s\n", (link.link_duplex == ETH_LINK_FULL_DUPLEX) ?
("full-duplex") : ("half-duplex"));
printf("Promiscuous mode: %s\n",
rte_eth_promiscuous_get(port_id) ? "enabled" : "disabled");
printf("Allmulticast mode: %s\n",
rte_eth_allmulticast_get(port_id) ? "enabled" : "disabled");
printf("Maximum number of MAC addresses: %u\n",
(unsigned int)(port->dev_info.max_mac_addrs));
printf("Maximum number of MAC addresses of hash filtering: %u\n",
(unsigned int)(port->dev_info.max_hash_mac_addrs));
vlan_offload = rte_eth_dev_get_vlan_offload(port_id);
if (vlan_offload >= 0){
printf("VLAN offload: \n");
if (vlan_offload & ETH_VLAN_STRIP_OFFLOAD)
printf(" strip on \n");
else
printf(" strip off \n");
if (vlan_offload & ETH_VLAN_FILTER_OFFLOAD)
printf(" filter on \n");
else
printf(" filter off \n");
if (vlan_offload & ETH_VLAN_EXTEND_OFFLOAD)
printf(" qinq(extend) on \n");
else
printf(" qinq(extend) off \n");
}
memset(&dev_info, 0, sizeof(dev_info));
rte_eth_dev_info_get(port_id, &dev_info);
if (dev_info.hash_key_size > 0)
printf("Hash key size in bytes: %u\n", dev_info.hash_key_size);
if (dev_info.reta_size > 0)
printf("Redirection table size: %u\n", dev_info.reta_size);
if (!dev_info.flow_type_rss_offloads)
printf("No flow type is supported.\n");
else {
uint16_t i;
char *p;
printf("Supported flow types:\n");
for (i = RTE_ETH_FLOW_UNKNOWN + 1; i < RTE_ETH_FLOW_MAX;
i++) {
if (!(dev_info.flow_type_rss_offloads & (1ULL << i)))
continue;
p = flowtype_to_str(i);
printf(" %s\n", (p ? p : "unknown"));
}
}
printf("Max possible RX queues: %u\n", dev_info.max_rx_queues);
printf("Max possible number of RXDs per queue: %hu\n",
dev_info.rx_desc_lim.nb_max);
printf("Min possible number of RXDs per queue: %hu\n",
dev_info.rx_desc_lim.nb_min);
printf("RXDs number alignment: %hu\n", dev_info.rx_desc_lim.nb_align);
printf("Max possible TX queues: %u\n", dev_info.max_tx_queues);
printf("Max possible number of TXDs per queue: %hu\n",
dev_info.tx_desc_lim.nb_max);
printf("Min possible number of TXDs per queue: %hu\n",
dev_info.tx_desc_lim.nb_min);
printf("TXDs number alignment: %hu\n", dev_info.tx_desc_lim.nb_align);
}
int
port_id_is_invalid(portid_t port_id, enum print_warning warning)
{
if (port_id == (portid_t)RTE_PORT_ALL)
return 0;
if (port_id < RTE_MAX_ETHPORTS && ports[port_id].enabled)
return 0;
if (warning == ENABLED_WARN)
printf("Invalid port %d\n", port_id);
return 1;
}
static int
vlan_id_is_invalid(uint16_t vlan_id)
{
if (vlan_id < 4096)
return 0;
printf("Invalid vlan_id %d (must be < 4096)\n", vlan_id);
return 1;
}
static int
port_reg_off_is_invalid(portid_t port_id, uint32_t reg_off)
{
uint64_t pci_len;
if (reg_off & 0x3) {
printf("Port register offset 0x%X not aligned on a 4-byte "
"boundary\n",
(unsigned)reg_off);
return 1;
}
pci_len = ports[port_id].dev_info.pci_dev->mem_resource[0].len;
if (reg_off >= pci_len) {
printf("Port %d: register offset %u (0x%X) out of port PCI "
"resource (length=%"PRIu64")\n",
port_id, (unsigned)reg_off, (unsigned)reg_off, pci_len);
return 1;
}
return 0;
}
static int
reg_bit_pos_is_invalid(uint8_t bit_pos)
{
if (bit_pos <= 31)
return 0;
printf("Invalid bit position %d (must be <= 31)\n", bit_pos);
return 1;
}
#define display_port_and_reg_off(port_id, reg_off) \
printf("port %d PCI register at offset 0x%X: ", (port_id), (reg_off))
static inline void
display_port_reg_value(portid_t port_id, uint32_t reg_off, uint32_t reg_v)
{
display_port_and_reg_off(port_id, (unsigned)reg_off);
printf("0x%08X (%u)\n", (unsigned)reg_v, (unsigned)reg_v);
}
void
port_reg_bit_display(portid_t port_id, uint32_t reg_off, uint8_t bit_x)
{
uint32_t reg_v;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
if (port_reg_off_is_invalid(port_id, reg_off))
return;
if (reg_bit_pos_is_invalid(bit_x))
return;
reg_v = port_id_pci_reg_read(port_id, reg_off);
display_port_and_reg_off(port_id, (unsigned)reg_off);
printf("bit %d=%d\n", bit_x, (int) ((reg_v & (1 << bit_x)) >> bit_x));
}
void
port_reg_bit_field_display(portid_t port_id, uint32_t reg_off,
uint8_t bit1_pos, uint8_t bit2_pos)
{
uint32_t reg_v;
uint8_t l_bit;
uint8_t h_bit;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
if (port_reg_off_is_invalid(port_id, reg_off))
return;
if (reg_bit_pos_is_invalid(bit1_pos))
return;
if (reg_bit_pos_is_invalid(bit2_pos))
return;
if (bit1_pos > bit2_pos)
l_bit = bit2_pos, h_bit = bit1_pos;
else
l_bit = bit1_pos, h_bit = bit2_pos;
reg_v = port_id_pci_reg_read(port_id, reg_off);
reg_v >>= l_bit;
if (h_bit < 31)
reg_v &= ((1 << (h_bit - l_bit + 1)) - 1);
display_port_and_reg_off(port_id, (unsigned)reg_off);
printf("bits[%d, %d]=0x%0*X (%u)\n", l_bit, h_bit,
((h_bit - l_bit) / 4) + 1, (unsigned)reg_v, (unsigned)reg_v);
}
void
port_reg_display(portid_t port_id, uint32_t reg_off)
{
uint32_t reg_v;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
if (port_reg_off_is_invalid(port_id, reg_off))
return;
reg_v = port_id_pci_reg_read(port_id, reg_off);
display_port_reg_value(port_id, reg_off, reg_v);
}
void
port_reg_bit_set(portid_t port_id, uint32_t reg_off, uint8_t bit_pos,
uint8_t bit_v)
{
uint32_t reg_v;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
if (port_reg_off_is_invalid(port_id, reg_off))
return;
if (reg_bit_pos_is_invalid(bit_pos))
return;
if (bit_v > 1) {
printf("Invalid bit value %d (must be 0 or 1)\n", (int) bit_v);
return;
}
reg_v = port_id_pci_reg_read(port_id, reg_off);
if (bit_v == 0)
reg_v &= ~(1 << bit_pos);
else
reg_v |= (1 << bit_pos);
port_id_pci_reg_write(port_id, reg_off, reg_v);
display_port_reg_value(port_id, reg_off, reg_v);
}
void
port_reg_bit_field_set(portid_t port_id, uint32_t reg_off,
uint8_t bit1_pos, uint8_t bit2_pos, uint32_t value)
{
uint32_t max_v;
uint32_t reg_v;
uint8_t l_bit;
uint8_t h_bit;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
if (port_reg_off_is_invalid(port_id, reg_off))
return;
if (reg_bit_pos_is_invalid(bit1_pos))
return;
if (reg_bit_pos_is_invalid(bit2_pos))
return;
if (bit1_pos > bit2_pos)
l_bit = bit2_pos, h_bit = bit1_pos;
else
l_bit = bit1_pos, h_bit = bit2_pos;
if ((h_bit - l_bit) < 31)
max_v = (1 << (h_bit - l_bit + 1)) - 1;
else
max_v = 0xFFFFFFFF;
if (value > max_v) {
printf("Invalid value %u (0x%x) must be < %u (0x%x)\n",
(unsigned)value, (unsigned)value,
(unsigned)max_v, (unsigned)max_v);
return;
}
reg_v = port_id_pci_reg_read(port_id, reg_off);
reg_v &= ~(max_v << l_bit); /* Keep unchanged bits */
reg_v |= (value << l_bit); /* Set changed bits */
port_id_pci_reg_write(port_id, reg_off, reg_v);
display_port_reg_value(port_id, reg_off, reg_v);
}
void
port_reg_set(portid_t port_id, uint32_t reg_off, uint32_t reg_v)
{
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
if (port_reg_off_is_invalid(port_id, reg_off))
return;
port_id_pci_reg_write(port_id, reg_off, reg_v);
display_port_reg_value(port_id, reg_off, reg_v);
}
void
port_mtu_set(portid_t port_id, uint16_t mtu)
{
int diag;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
diag = rte_eth_dev_set_mtu(port_id, mtu);
if (diag == 0)
return;
printf("Set MTU failed. diag=%d\n", diag);
}
/*
* RX/TX ring descriptors display functions.
*/
int
rx_queue_id_is_invalid(queueid_t rxq_id)
{
if (rxq_id < nb_rxq)
return 0;
printf("Invalid RX queue %d (must be < nb_rxq=%d)\n", rxq_id, nb_rxq);
return 1;
}
int
tx_queue_id_is_invalid(queueid_t txq_id)
{
if (txq_id < nb_txq)
return 0;
printf("Invalid TX queue %d (must be < nb_rxq=%d)\n", txq_id, nb_txq);
return 1;
}
static int
rx_desc_id_is_invalid(uint16_t rxdesc_id)
{
if (rxdesc_id < nb_rxd)
return 0;
printf("Invalid RX descriptor %d (must be < nb_rxd=%d)\n",
rxdesc_id, nb_rxd);
return 1;
}
static int
tx_desc_id_is_invalid(uint16_t txdesc_id)
{
if (txdesc_id < nb_txd)
return 0;
printf("Invalid TX descriptor %d (must be < nb_txd=%d)\n",
txdesc_id, nb_txd);
return 1;
}
static const struct rte_memzone *
ring_dma_zone_lookup(const char *ring_name, uint8_t port_id, uint16_t q_id)
{
char mz_name[RTE_MEMZONE_NAMESIZE];
const struct rte_memzone *mz;
snprintf(mz_name, sizeof(mz_name), "%s_%s_%d_%d",
ports[port_id].dev_info.driver_name, ring_name, port_id, q_id);
mz = rte_memzone_lookup(mz_name);
if (mz == NULL)
printf("%s ring memory zoneof (port %d, queue %d) not"
"found (zone name = %s\n",
ring_name, port_id, q_id, mz_name);
return mz;
}
union igb_ring_dword {
uint64_t dword;
struct {
#if RTE_BYTE_ORDER == RTE_BIG_ENDIAN
uint32_t lo;
uint32_t hi;
#else
uint32_t hi;
uint32_t lo;
#endif
} words;
};
struct igb_ring_desc_32_bytes {
union igb_ring_dword lo_dword;
union igb_ring_dword hi_dword;
union igb_ring_dword resv1;
union igb_ring_dword resv2;
};
struct igb_ring_desc_16_bytes {
union igb_ring_dword lo_dword;
union igb_ring_dword hi_dword;
};
static void
ring_rxd_display_dword(union igb_ring_dword dword)
{
printf(" 0x%08X - 0x%08X\n", (unsigned)dword.words.lo,
(unsigned)dword.words.hi);
}
static void
ring_rx_descriptor_display(const struct rte_memzone *ring_mz,
#ifndef RTE_LIBRTE_I40E_16BYTE_RX_DESC
uint8_t port_id,
#else
__rte_unused uint8_t port_id,
#endif
uint16_t desc_id)
{
struct igb_ring_desc_16_bytes *ring =
(struct igb_ring_desc_16_bytes *)ring_mz->addr;
#ifndef RTE_LIBRTE_I40E_16BYTE_RX_DESC
struct rte_eth_dev_info dev_info;
memset(&dev_info, 0, sizeof(dev_info));
rte_eth_dev_info_get(port_id, &dev_info);
if (strstr(dev_info.driver_name, "i40e") != NULL) {
/* 32 bytes RX descriptor, i40e only */
struct igb_ring_desc_32_bytes *ring =
(struct igb_ring_desc_32_bytes *)ring_mz->addr;
ring[desc_id].lo_dword.dword =
rte_le_to_cpu_64(ring[desc_id].lo_dword.dword);
ring_rxd_display_dword(ring[desc_id].lo_dword);
ring[desc_id].hi_dword.dword =
rte_le_to_cpu_64(ring[desc_id].hi_dword.dword);
ring_rxd_display_dword(ring[desc_id].hi_dword);
ring[desc_id].resv1.dword =
rte_le_to_cpu_64(ring[desc_id].resv1.dword);
ring_rxd_display_dword(ring[desc_id].resv1);
ring[desc_id].resv2.dword =
rte_le_to_cpu_64(ring[desc_id].resv2.dword);
ring_rxd_display_dword(ring[desc_id].resv2);
return;
}
#endif
/* 16 bytes RX descriptor */
ring[desc_id].lo_dword.dword =
rte_le_to_cpu_64(ring[desc_id].lo_dword.dword);
ring_rxd_display_dword(ring[desc_id].lo_dword);
ring[desc_id].hi_dword.dword =
rte_le_to_cpu_64(ring[desc_id].hi_dword.dword);
ring_rxd_display_dword(ring[desc_id].hi_dword);
}
static void
ring_tx_descriptor_display(const struct rte_memzone *ring_mz, uint16_t desc_id)
{
struct igb_ring_desc_16_bytes *ring;
struct igb_ring_desc_16_bytes txd;
ring = (struct igb_ring_desc_16_bytes *)ring_mz->addr;
txd.lo_dword.dword = rte_le_to_cpu_64(ring[desc_id].lo_dword.dword);
txd.hi_dword.dword = rte_le_to_cpu_64(ring[desc_id].hi_dword.dword);
printf(" 0x%08X - 0x%08X / 0x%08X - 0x%08X\n",
(unsigned)txd.lo_dword.words.lo,
(unsigned)txd.lo_dword.words.hi,
(unsigned)txd.hi_dword.words.lo,
(unsigned)txd.hi_dword.words.hi);
}
void
rx_ring_desc_display(portid_t port_id, queueid_t rxq_id, uint16_t rxd_id)
{
const struct rte_memzone *rx_mz;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
if (rx_queue_id_is_invalid(rxq_id))
return;
if (rx_desc_id_is_invalid(rxd_id))
return;
rx_mz = ring_dma_zone_lookup("rx_ring", port_id, rxq_id);
if (rx_mz == NULL)
return;
ring_rx_descriptor_display(rx_mz, port_id, rxd_id);
}
void
tx_ring_desc_display(portid_t port_id, queueid_t txq_id, uint16_t txd_id)
{
const struct rte_memzone *tx_mz;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
if (tx_queue_id_is_invalid(txq_id))
return;
if (tx_desc_id_is_invalid(txd_id))
return;
tx_mz = ring_dma_zone_lookup("tx_ring", port_id, txq_id);
if (tx_mz == NULL)
return;
ring_tx_descriptor_display(tx_mz, txd_id);
}
void
fwd_lcores_config_display(void)
{
lcoreid_t lc_id;
printf("List of forwarding lcores:");
for (lc_id = 0; lc_id < nb_cfg_lcores; lc_id++)
printf(" %2u", fwd_lcores_cpuids[lc_id]);
printf("\n");
}
void
rxtx_config_display(void)
{
printf(" %s packet forwarding - CRC stripping %s - "
"packets/burst=%d\n", cur_fwd_eng->fwd_mode_name,
rx_mode.hw_strip_crc ? "enabled" : "disabled",
nb_pkt_per_burst);
if (cur_fwd_eng == &tx_only_engine)
printf(" packet len=%u - nb packet segments=%d\n",
(unsigned)tx_pkt_length, (int) tx_pkt_nb_segs);
struct rte_eth_rxconf *rx_conf = &ports[0].rx_conf;
struct rte_eth_txconf *tx_conf = &ports[0].tx_conf;
printf(" nb forwarding cores=%d - nb forwarding ports=%d\n",
nb_fwd_lcores, nb_fwd_ports);
printf(" RX queues=%d - RX desc=%d - RX free threshold=%d\n",
nb_rxq, nb_rxd, rx_conf->rx_free_thresh);
printf(" RX threshold registers: pthresh=%d hthresh=%d wthresh=%d\n",
rx_conf->rx_thresh.pthresh, rx_conf->rx_thresh.hthresh,
rx_conf->rx_thresh.wthresh);
printf(" TX queues=%d - TX desc=%d - TX free threshold=%d\n",
nb_txq, nb_txd, tx_conf->tx_free_thresh);
printf(" TX threshold registers: pthresh=%d hthresh=%d wthresh=%d\n",
tx_conf->tx_thresh.pthresh, tx_conf->tx_thresh.hthresh,
tx_conf->tx_thresh.wthresh);
printf(" TX RS bit threshold=%d - TXQ flags=0x%"PRIx32"\n",
tx_conf->tx_rs_thresh, tx_conf->txq_flags);
}
void
port_rss_reta_info(portid_t port_id,
struct rte_eth_rss_reta_entry64 *reta_conf,
uint16_t nb_entries)
{
uint16_t i, idx, shift;
int ret;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
ret = rte_eth_dev_rss_reta_query(port_id, reta_conf, nb_entries);
if (ret != 0) {
printf("Failed to get RSS RETA info, return code = %d\n", ret);
return;
}
for (i = 0; i < nb_entries; i++) {
idx = i / RTE_RETA_GROUP_SIZE;
shift = i % RTE_RETA_GROUP_SIZE;
if (!(reta_conf[idx].mask & (1ULL << shift)))
continue;
printf("RSS RETA configuration: hash index=%u, queue=%u\n",
i, reta_conf[idx].reta[shift]);
}
}
/*
* Displays the RSS hash functions of a port, and, optionaly, the RSS hash
* key of the port.
*/
void
port_rss_hash_conf_show(portid_t port_id, char rss_info[], int show_rss_key)
{
struct rte_eth_rss_conf rss_conf;
uint8_t rss_key[10 * 4] = "";
uint64_t rss_hf;
uint8_t i;
int diag;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
rss_conf.rss_hf = 0;
for (i = 0; i < RTE_DIM(rss_type_table); i++) {
if (!strcmp(rss_info, rss_type_table[i].str))
rss_conf.rss_hf = rss_type_table[i].rss_type;
}
/* Get RSS hash key if asked to display it */
rss_conf.rss_key = (show_rss_key) ? rss_key : NULL;
rss_conf.rss_key_len = sizeof(rss_key);
diag = rte_eth_dev_rss_hash_conf_get(port_id, &rss_conf);
if (diag != 0) {
switch (diag) {
case -ENODEV:
printf("port index %d invalid\n", port_id);
break;
case -ENOTSUP:
printf("operation not supported by device\n");
break;
default:
printf("operation failed - diag=%d\n", diag);
break;
}
return;
}
rss_hf = rss_conf.rss_hf;
if (rss_hf == 0) {
printf("RSS disabled\n");
return;
}
printf("RSS functions:\n ");
for (i = 0; i < RTE_DIM(rss_type_table); i++) {
if (rss_hf & rss_type_table[i].rss_type)
printf("%s ", rss_type_table[i].str);
}
printf("\n");
if (!show_rss_key)
return;
printf("RSS key:\n");
for (i = 0; i < sizeof(rss_key); i++)
printf("%02X", rss_key[i]);
printf("\n");
}
void
port_rss_hash_key_update(portid_t port_id, char rss_type[], uint8_t *hash_key,
uint hash_key_len)
{
struct rte_eth_rss_conf rss_conf;
int diag;
unsigned int i;
rss_conf.rss_key = NULL;
rss_conf.rss_key_len = hash_key_len;
rss_conf.rss_hf = 0;
for (i = 0; i < RTE_DIM(rss_type_table); i++) {
if (!strcmp(rss_type_table[i].str, rss_type))
rss_conf.rss_hf = rss_type_table[i].rss_type;
}
diag = rte_eth_dev_rss_hash_conf_get(port_id, &rss_conf);
if (diag == 0) {
rss_conf.rss_key = hash_key;
diag = rte_eth_dev_rss_hash_update(port_id, &rss_conf);
}
if (diag == 0)
return;
switch (diag) {
case -ENODEV:
printf("port index %d invalid\n", port_id);
break;
case -ENOTSUP:
printf("operation not supported by device\n");
break;
default:
printf("operation failed - diag=%d\n", diag);
break;
}
}
/*
* Setup forwarding configuration for each logical core.
*/
static void
setup_fwd_config_of_each_lcore(struct fwd_config *cfg)
{
streamid_t nb_fs_per_lcore;
streamid_t nb_fs;
streamid_t sm_id;
lcoreid_t nb_extra;
lcoreid_t nb_fc;
lcoreid_t nb_lc;
lcoreid_t lc_id;
nb_fs = cfg->nb_fwd_streams;
nb_fc = cfg->nb_fwd_lcores;
if (nb_fs <= nb_fc) {
nb_fs_per_lcore = 1;
nb_extra = 0;
} else {
nb_fs_per_lcore = (streamid_t) (nb_fs / nb_fc);
nb_extra = (lcoreid_t) (nb_fs % nb_fc);
}
nb_lc = (lcoreid_t) (nb_fc - nb_extra);
sm_id = 0;
for (lc_id = 0; lc_id < nb_lc; lc_id++) {
fwd_lcores[lc_id]->stream_idx = sm_id;
fwd_lcores[lc_id]->stream_nb = nb_fs_per_lcore;
sm_id = (streamid_t) (sm_id + nb_fs_per_lcore);
}
/*
* Assign extra remaining streams, if any.
*/
nb_fs_per_lcore = (streamid_t) (nb_fs_per_lcore + 1);
for (lc_id = 0; lc_id < nb_extra; lc_id++) {
fwd_lcores[nb_lc + lc_id]->stream_idx = sm_id;
fwd_lcores[nb_lc + lc_id]->stream_nb = nb_fs_per_lcore;
sm_id = (streamid_t) (sm_id + nb_fs_per_lcore);
}
}
static void
simple_fwd_config_setup(void)
{
portid_t i;
portid_t j;
portid_t inc = 2;
if (port_topology == PORT_TOPOLOGY_CHAINED ||
port_topology == PORT_TOPOLOGY_LOOP) {
inc = 1;
} else if (nb_fwd_ports % 2) {
printf("\nWarning! Cannot handle an odd number of ports "
"with the current port topology. Configuration "
"must be changed to have an even number of ports, "
"or relaunch application with "
"--port-topology=chained\n\n");
}
cur_fwd_config.nb_fwd_ports = (portid_t) nb_fwd_ports;
cur_fwd_config.nb_fwd_streams =
(streamid_t) cur_fwd_config.nb_fwd_ports;
/* reinitialize forwarding streams */
init_fwd_streams();
/*
* In the simple forwarding test, the number of forwarding cores
* must be lower or equal to the number of forwarding ports.
*/
cur_fwd_config.nb_fwd_lcores = (lcoreid_t) nb_fwd_lcores;
if (cur_fwd_config.nb_fwd_lcores > cur_fwd_config.nb_fwd_ports)
cur_fwd_config.nb_fwd_lcores =
(lcoreid_t) cur_fwd_config.nb_fwd_ports;
setup_fwd_config_of_each_lcore(&cur_fwd_config);
for (i = 0; i < cur_fwd_config.nb_fwd_ports; i = (portid_t) (i + inc)) {
if (port_topology != PORT_TOPOLOGY_LOOP)
j = (portid_t) ((i + 1) % cur_fwd_config.nb_fwd_ports);
else
j = i;
fwd_streams[i]->rx_port = fwd_ports_ids[i];
fwd_streams[i]->rx_queue = 0;
fwd_streams[i]->tx_port = fwd_ports_ids[j];
fwd_streams[i]->tx_queue = 0;
fwd_streams[i]->peer_addr = j;
if (port_topology == PORT_TOPOLOGY_PAIRED) {
fwd_streams[j]->rx_port = fwd_ports_ids[j];
fwd_streams[j]->rx_queue = 0;
fwd_streams[j]->tx_port = fwd_ports_ids[i];
fwd_streams[j]->tx_queue = 0;
fwd_streams[j]->peer_addr = i;
}
}
}
/**
* For the RSS forwarding test, each core is assigned on every port a transmit
* queue whose index is the index of the core itself. This approach limits the
* maximumm number of processing cores of the RSS test to the maximum number of
* TX queues supported by the devices.
*
* Each core is assigned a single stream, each stream being composed of
* a RX queue to poll on a RX port for input messages, associated with
* a TX queue of a TX port where to send forwarded packets.
* All packets received on the RX queue of index "RxQj" of the RX port "RxPi"
* are sent on the TX queue "TxQl" of the TX port "TxPk" according to the two
* following rules:
* - TxPk = (RxPi + 1) if RxPi is even, (RxPi - 1) if RxPi is odd
* - TxQl = RxQj
*/
static void
rss_fwd_config_setup(void)
{
portid_t rxp;
portid_t txp;
queueid_t rxq;
queueid_t nb_q;
lcoreid_t lc_id;
nb_q = nb_rxq;
if (nb_q > nb_txq)
nb_q = nb_txq;
cur_fwd_config.nb_fwd_lcores = (lcoreid_t) nb_fwd_lcores;
cur_fwd_config.nb_fwd_ports = nb_fwd_ports;
cur_fwd_config.nb_fwd_streams =
(streamid_t) (nb_q * cur_fwd_config.nb_fwd_ports);
if (cur_fwd_config.nb_fwd_streams > cur_fwd_config.nb_fwd_lcores)
cur_fwd_config.nb_fwd_streams =
(streamid_t)cur_fwd_config.nb_fwd_lcores;
else
cur_fwd_config.nb_fwd_lcores =
(lcoreid_t)cur_fwd_config.nb_fwd_streams;
/* reinitialize forwarding streams */
init_fwd_streams();
setup_fwd_config_of_each_lcore(&cur_fwd_config);
rxp = 0; rxq = 0;
for (lc_id = 0; lc_id < cur_fwd_config.nb_fwd_lcores; lc_id++) {
struct fwd_stream *fs;
fs = fwd_streams[lc_id];
if ((rxp & 0x1) == 0)
txp = (portid_t) (rxp + 1);
else
txp = (portid_t) (rxp - 1);
/*
* if we are in loopback, simply send stuff out through the
* ingress port
*/
if (port_topology == PORT_TOPOLOGY_LOOP)
txp = rxp;
fs->rx_port = fwd_ports_ids[rxp];
fs->rx_queue = rxq;
fs->tx_port = fwd_ports_ids[txp];
fs->tx_queue = rxq;
fs->peer_addr = fs->tx_port;
rxq = (queueid_t) (rxq + 1);
if (rxq < nb_q)
continue;
/*
* rxq == nb_q
* Restart from RX queue 0 on next RX port
*/
rxq = 0;
if (numa_support && (nb_fwd_ports <= (nb_ports >> 1)))
rxp = (portid_t)
(rxp + ((nb_ports >> 1) / nb_fwd_ports));
else
rxp = (portid_t) (rxp + 1);
}
}
/**
* For the DCB forwarding test, each core is assigned on each traffic class.
*
* Each core is assigned a multi-stream, each stream being composed of
* a RX queue to poll on a RX port for input messages, associated with
* a TX queue of a TX port where to send forwarded packets. All RX and
* TX queues are mapping to the same traffic class.
* If VMDQ and DCB co-exist, each traffic class on different POOLs share
* the same core
*/
static void
dcb_fwd_config_setup(void)
{
struct rte_eth_dcb_info rxp_dcb_info, txp_dcb_info;
portid_t txp, rxp = 0;
queueid_t txq, rxq = 0;
lcoreid_t lc_id;
uint16_t nb_rx_queue, nb_tx_queue;
uint16_t i, j, k, sm_id = 0;
uint8_t tc = 0;
cur_fwd_config.nb_fwd_lcores = (lcoreid_t) nb_fwd_lcores;
cur_fwd_config.nb_fwd_ports = nb_fwd_ports;
cur_fwd_config.nb_fwd_streams =
(streamid_t) (nb_rxq * cur_fwd_config.nb_fwd_ports);
/* reinitialize forwarding streams */
init_fwd_streams();
sm_id = 0;
txp = 1;
/* get the dcb info on the first RX and TX ports */
(void)rte_eth_dev_get_dcb_info(fwd_ports_ids[rxp], &rxp_dcb_info);
(void)rte_eth_dev_get_dcb_info(fwd_ports_ids[txp], &txp_dcb_info);
for (lc_id = 0; lc_id < cur_fwd_config.nb_fwd_lcores; lc_id++) {
fwd_lcores[lc_id]->stream_nb = 0;
fwd_lcores[lc_id]->stream_idx = sm_id;
for (i = 0; i < ETH_MAX_VMDQ_POOL; i++) {
/* if the nb_queue is zero, means this tc is
* not enabled on the POOL
*/
if (rxp_dcb_info.tc_queue.tc_rxq[i][tc].nb_queue == 0)
break;
k = fwd_lcores[lc_id]->stream_nb +
fwd_lcores[lc_id]->stream_idx;
rxq = rxp_dcb_info.tc_queue.tc_rxq[i][tc].base;
txq = txp_dcb_info.tc_queue.tc_txq[i][tc].base;
nb_rx_queue = txp_dcb_info.tc_queue.tc_rxq[i][tc].nb_queue;
nb_tx_queue = txp_dcb_info.tc_queue.tc_txq[i][tc].nb_queue;
for (j = 0; j < nb_rx_queue; j++) {
struct fwd_stream *fs;
fs = fwd_streams[k + j];
fs->rx_port = fwd_ports_ids[rxp];
fs->rx_queue = rxq + j;
fs->tx_port = fwd_ports_ids[txp];
fs->tx_queue = txq + j % nb_tx_queue;
fs->peer_addr = fs->tx_port;
}
fwd_lcores[lc_id]->stream_nb +=
rxp_dcb_info.tc_queue.tc_rxq[i][tc].nb_queue;
}
sm_id = (streamid_t) (sm_id + fwd_lcores[lc_id]->stream_nb);
tc++;
if (tc < rxp_dcb_info.nb_tcs)
continue;
/* Restart from TC 0 on next RX port */
tc = 0;
if (numa_support && (nb_fwd_ports <= (nb_ports >> 1)))
rxp = (portid_t)
(rxp + ((nb_ports >> 1) / nb_fwd_ports));
else
rxp++;
if (rxp >= nb_fwd_ports)
return;
/* get the dcb information on next RX and TX ports */
if ((rxp & 0x1) == 0)
txp = (portid_t) (rxp + 1);
else
txp = (portid_t) (rxp - 1);
rte_eth_dev_get_dcb_info(fwd_ports_ids[rxp], &rxp_dcb_info);
rte_eth_dev_get_dcb_info(fwd_ports_ids[txp], &txp_dcb_info);
}
}
static void
icmp_echo_config_setup(void)
{
portid_t rxp;
queueid_t rxq;
lcoreid_t lc_id;
uint16_t sm_id;
if ((nb_txq * nb_fwd_ports) < nb_fwd_lcores)
cur_fwd_config.nb_fwd_lcores = (lcoreid_t)
(nb_txq * nb_fwd_ports);
else
cur_fwd_config.nb_fwd_lcores = (lcoreid_t) nb_fwd_lcores;
cur_fwd_config.nb_fwd_ports = nb_fwd_ports;
cur_fwd_config.nb_fwd_streams =
(streamid_t) (nb_rxq * cur_fwd_config.nb_fwd_ports);
if (cur_fwd_config.nb_fwd_streams < cur_fwd_config.nb_fwd_lcores)
cur_fwd_config.nb_fwd_lcores =
(lcoreid_t)cur_fwd_config.nb_fwd_streams;
if (verbose_level > 0) {
printf("%s fwd_cores=%d fwd_ports=%d fwd_streams=%d\n",
__FUNCTION__,
cur_fwd_config.nb_fwd_lcores,
cur_fwd_config.nb_fwd_ports,
cur_fwd_config.nb_fwd_streams);
}
/* reinitialize forwarding streams */
init_fwd_streams();
setup_fwd_config_of_each_lcore(&cur_fwd_config);
rxp = 0; rxq = 0;
for (lc_id = 0; lc_id < cur_fwd_config.nb_fwd_lcores; lc_id++) {
if (verbose_level > 0)
printf(" core=%d: \n", lc_id);
for (sm_id = 0; sm_id < fwd_lcores[lc_id]->stream_nb; sm_id++) {
struct fwd_stream *fs;
fs = fwd_streams[fwd_lcores[lc_id]->stream_idx + sm_id];
fs->rx_port = fwd_ports_ids[rxp];
fs->rx_queue = rxq;
fs->tx_port = fs->rx_port;
fs->tx_queue = rxq;
fs->peer_addr = fs->tx_port;
if (verbose_level > 0)
printf(" stream=%d port=%d rxq=%d txq=%d\n",
sm_id, fs->rx_port, fs->rx_queue,
fs->tx_queue);
rxq = (queueid_t) (rxq + 1);
if (rxq == nb_rxq) {
rxq = 0;
rxp = (portid_t) (rxp + 1);
}
}
}
}
void
fwd_config_setup(void)
{
cur_fwd_config.fwd_eng = cur_fwd_eng;
if (strcmp(cur_fwd_eng->fwd_mode_name, "icmpecho") == 0) {
icmp_echo_config_setup();
return;
}
if ((nb_rxq > 1) && (nb_txq > 1)){
if (dcb_config)
dcb_fwd_config_setup();
else
rss_fwd_config_setup();
}
else
simple_fwd_config_setup();
}
static void
pkt_fwd_config_display(struct fwd_config *cfg)
{
struct fwd_stream *fs;
lcoreid_t lc_id;
streamid_t sm_id;
printf("%s packet forwarding - ports=%d - cores=%d - streams=%d - "
"NUMA support %s, MP over anonymous pages %s\n",
cfg->fwd_eng->fwd_mode_name,
cfg->nb_fwd_ports, cfg->nb_fwd_lcores, cfg->nb_fwd_streams,
numa_support == 1 ? "enabled" : "disabled",
mp_anon != 0 ? "enabled" : "disabled");
if (strcmp(cfg->fwd_eng->fwd_mode_name, "mac_retry") == 0)
printf("TX retry num: %u, delay between TX retries: %uus\n",
burst_tx_retry_num, burst_tx_delay_time);
for (lc_id = 0; lc_id < cfg->nb_fwd_lcores; lc_id++) {
printf("Logical Core %u (socket %u) forwards packets on "
"%d streams:",
fwd_lcores_cpuids[lc_id],
rte_lcore_to_socket_id(fwd_lcores_cpuids[lc_id]),
fwd_lcores[lc_id]->stream_nb);
for (sm_id = 0; sm_id < fwd_lcores[lc_id]->stream_nb; sm_id++) {
fs = fwd_streams[fwd_lcores[lc_id]->stream_idx + sm_id];
printf("\n RX P=%d/Q=%d (socket %u) -> TX "
"P=%d/Q=%d (socket %u) ",
fs->rx_port, fs->rx_queue,
ports[fs->rx_port].socket_id,
fs->tx_port, fs->tx_queue,
ports[fs->tx_port].socket_id);
print_ethaddr("peer=",
&peer_eth_addrs[fs->peer_addr]);
}
printf("\n");
}
printf("\n");
}
void
fwd_config_display(void)
{
fwd_config_setup();
pkt_fwd_config_display(&cur_fwd_config);
}
int
set_fwd_lcores_list(unsigned int *lcorelist, unsigned int nb_lc)
{
unsigned int i;
unsigned int lcore_cpuid;
int record_now;
record_now = 0;
again:
for (i = 0; i < nb_lc; i++) {
lcore_cpuid = lcorelist[i];
if (! rte_lcore_is_enabled(lcore_cpuid)) {
printf("lcore %u not enabled\n", lcore_cpuid);
return -1;
}
if (lcore_cpuid == rte_get_master_lcore()) {
printf("lcore %u cannot be masked on for running "
"packet forwarding, which is the master lcore "
"and reserved for command line parsing only\n",
lcore_cpuid);
return -1;
}
if (record_now)
fwd_lcores_cpuids[i] = lcore_cpuid;
}
if (record_now == 0) {
record_now = 1;
goto again;
}
nb_cfg_lcores = (lcoreid_t) nb_lc;
if (nb_fwd_lcores != (lcoreid_t) nb_lc) {
printf("previous number of forwarding cores %u - changed to "
"number of configured cores %u\n",
(unsigned int) nb_fwd_lcores, nb_lc);
nb_fwd_lcores = (lcoreid_t) nb_lc;
}
return 0;
}
int
set_fwd_lcores_mask(uint64_t lcoremask)
{
unsigned int lcorelist[64];
unsigned int nb_lc;
unsigned int i;
if (lcoremask == 0) {
printf("Invalid NULL mask of cores\n");
return -1;
}
nb_lc = 0;
for (i = 0; i < 64; i++) {
if (! ((uint64_t)(1ULL << i) & lcoremask))
continue;
lcorelist[nb_lc++] = i;
}
return set_fwd_lcores_list(lcorelist, nb_lc);
}
void
set_fwd_lcores_number(uint16_t nb_lc)
{
if (nb_lc > nb_cfg_lcores) {
printf("nb fwd cores %u > %u (max. number of configured "
"lcores) - ignored\n",
(unsigned int) nb_lc, (unsigned int) nb_cfg_lcores);
return;
}
nb_fwd_lcores = (lcoreid_t) nb_lc;
printf("Number of forwarding cores set to %u\n",
(unsigned int) nb_fwd_lcores);
}
void
set_fwd_ports_list(unsigned int *portlist, unsigned int nb_pt)
{
unsigned int i;
portid_t port_id;
int record_now;
record_now = 0;
again:
for (i = 0; i < nb_pt; i++) {
port_id = (portid_t) portlist[i];
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
if (record_now)
fwd_ports_ids[i] = port_id;
}
if (record_now == 0) {
record_now = 1;
goto again;
}
nb_cfg_ports = (portid_t) nb_pt;
if (nb_fwd_ports != (portid_t) nb_pt) {
printf("previous number of forwarding ports %u - changed to "
"number of configured ports %u\n",
(unsigned int) nb_fwd_ports, nb_pt);
nb_fwd_ports = (portid_t) nb_pt;
}
}
void
set_fwd_ports_mask(uint64_t portmask)
{
unsigned int portlist[64];
unsigned int nb_pt;
unsigned int i;
if (portmask == 0) {
printf("Invalid NULL mask of ports\n");
return;
}
nb_pt = 0;
for (i = 0; i < (unsigned)RTE_MIN(64, RTE_MAX_ETHPORTS); i++) {
if (! ((uint64_t)(1ULL << i) & portmask))
continue;
portlist[nb_pt++] = i;
}
set_fwd_ports_list(portlist, nb_pt);
}
void
set_fwd_ports_number(uint16_t nb_pt)
{
if (nb_pt > nb_cfg_ports) {
printf("nb fwd ports %u > %u (number of configured "
"ports) - ignored\n",
(unsigned int) nb_pt, (unsigned int) nb_cfg_ports);
return;
}
nb_fwd_ports = (portid_t) nb_pt;
printf("Number of forwarding ports set to %u\n",
(unsigned int) nb_fwd_ports);
}
void
set_nb_pkt_per_burst(uint16_t nb)
{
if (nb > MAX_PKT_BURST) {
printf("nb pkt per burst: %u > %u (maximum packet per burst) "
" ignored\n",
(unsigned int) nb, (unsigned int) MAX_PKT_BURST);
return;
}
nb_pkt_per_burst = nb;
printf("Number of packets per burst set to %u\n",
(unsigned int) nb_pkt_per_burst);
}
static const char *
tx_split_get_name(enum tx_pkt_split split)
{
uint32_t i;
for (i = 0; i != RTE_DIM(tx_split_name); i++) {
if (tx_split_name[i].split == split)
return tx_split_name[i].name;
}
return NULL;
}
void
set_tx_pkt_split(const char *name)
{
uint32_t i;
for (i = 0; i != RTE_DIM(tx_split_name); i++) {
if (strcmp(tx_split_name[i].name, name) == 0) {
tx_pkt_split = tx_split_name[i].split;
return;
}
}
printf("unknown value: \"%s\"\n", name);
}
void
show_tx_pkt_segments(void)
{
uint32_t i, n;
const char *split;
n = tx_pkt_nb_segs;
split = tx_split_get_name(tx_pkt_split);
printf("Number of segments: %u\n", n);
printf("Segment sizes: ");
for (i = 0; i != n - 1; i++)
printf("%hu,", tx_pkt_seg_lengths[i]);
printf("%hu\n", tx_pkt_seg_lengths[i]);
printf("Split packet: %s\n", split);
}
void
set_tx_pkt_segments(unsigned *seg_lengths, unsigned nb_segs)
{
uint16_t tx_pkt_len;
unsigned i;
if (nb_segs >= (unsigned) nb_txd) {
printf("nb segments per TX packets=%u >= nb_txd=%u - ignored\n",
nb_segs, (unsigned int) nb_txd);
return;
}
/*
* Check that each segment length is greater or equal than
* the mbuf data sise.
* Check also that the total packet length is greater or equal than the
* size of an empty UDP/IP packet (sizeof(struct ether_hdr) + 20 + 8).
*/
tx_pkt_len = 0;
for (i = 0; i < nb_segs; i++) {
if (seg_lengths[i] > (unsigned) mbuf_data_size) {
printf("length[%u]=%u > mbuf_data_size=%u - give up\n",
i, seg_lengths[i], (unsigned) mbuf_data_size);
return;
}
tx_pkt_len = (uint16_t)(tx_pkt_len + seg_lengths[i]);
}
if (tx_pkt_len < (sizeof(struct ether_hdr) + 20 + 8)) {
printf("total packet length=%u < %d - give up\n",
(unsigned) tx_pkt_len,
(int)(sizeof(struct ether_hdr) + 20 + 8));
return;
}
for (i = 0; i < nb_segs; i++)
tx_pkt_seg_lengths[i] = (uint16_t) seg_lengths[i];
tx_pkt_length = tx_pkt_len;
tx_pkt_nb_segs = (uint8_t) nb_segs;
}
char*
list_pkt_forwarding_modes(void)
{
static char fwd_modes[128] = "";
const char *separator = "|";
struct fwd_engine *fwd_eng;
unsigned i = 0;
if (strlen (fwd_modes) == 0) {
while ((fwd_eng = fwd_engines[i++]) != NULL) {
strcat(fwd_modes, fwd_eng->fwd_mode_name);
strcat(fwd_modes, separator);
}
fwd_modes[strlen(fwd_modes) - strlen(separator)] = '\0';
}
return fwd_modes;
}
void
set_pkt_forwarding_mode(const char *fwd_mode_name)
{
struct fwd_engine *fwd_eng;
unsigned i;
i = 0;
while ((fwd_eng = fwd_engines[i]) != NULL) {
if (! strcmp(fwd_eng->fwd_mode_name, fwd_mode_name)) {
printf("Set %s packet forwarding mode\n",
fwd_mode_name);
cur_fwd_eng = fwd_eng;
return;
}
i++;
}
printf("Invalid %s packet forwarding mode\n", fwd_mode_name);
}
void
set_verbose_level(uint16_t vb_level)
{
printf("Change verbose level from %u to %u\n",
(unsigned int) verbose_level, (unsigned int) vb_level);
verbose_level = vb_level;
}
void
vlan_extend_set(portid_t port_id, int on)
{
int diag;
int vlan_offload;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
vlan_offload = rte_eth_dev_get_vlan_offload(port_id);
if (on)
vlan_offload |= ETH_VLAN_EXTEND_OFFLOAD;
else
vlan_offload &= ~ETH_VLAN_EXTEND_OFFLOAD;
diag = rte_eth_dev_set_vlan_offload(port_id, vlan_offload);
if (diag < 0)
printf("rx_vlan_extend_set(port_pi=%d, on=%d) failed "
"diag=%d\n", port_id, on, diag);
}
void
rx_vlan_strip_set(portid_t port_id, int on)
{
int diag;
int vlan_offload;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
vlan_offload = rte_eth_dev_get_vlan_offload(port_id);
if (on)
vlan_offload |= ETH_VLAN_STRIP_OFFLOAD;
else
vlan_offload &= ~ETH_VLAN_STRIP_OFFLOAD;
diag = rte_eth_dev_set_vlan_offload(port_id, vlan_offload);
if (diag < 0)
printf("rx_vlan_strip_set(port_pi=%d, on=%d) failed "
"diag=%d\n", port_id, on, diag);
}
void
rx_vlan_strip_set_on_queue(portid_t port_id, uint16_t queue_id, int on)
{
int diag;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
diag = rte_eth_dev_set_vlan_strip_on_queue(port_id, queue_id, on);
if (diag < 0)
printf("rx_vlan_strip_set_on_queue(port_pi=%d, queue_id=%d, on=%d) failed "
"diag=%d\n", port_id, queue_id, on, diag);
}
void
rx_vlan_filter_set(portid_t port_id, int on)
{
int diag;
int vlan_offload;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
vlan_offload = rte_eth_dev_get_vlan_offload(port_id);
if (on)
vlan_offload |= ETH_VLAN_FILTER_OFFLOAD;
else
vlan_offload &= ~ETH_VLAN_FILTER_OFFLOAD;
diag = rte_eth_dev_set_vlan_offload(port_id, vlan_offload);
if (diag < 0)
printf("rx_vlan_filter_set(port_pi=%d, on=%d) failed "
"diag=%d\n", port_id, on, diag);
}
int
rx_vft_set(portid_t port_id, uint16_t vlan_id, int on)
{
int diag;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return 1;
if (vlan_id_is_invalid(vlan_id))
return 1;
diag = rte_eth_dev_vlan_filter(port_id, vlan_id, on);
if (diag == 0)
return 0;
printf("rte_eth_dev_vlan_filter(port_pi=%d, vlan_id=%d, on=%d) failed "
"diag=%d\n",
port_id, vlan_id, on, diag);
return -1;
}
void
rx_vlan_all_filter_set(portid_t port_id, int on)
{
uint16_t vlan_id;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
for (vlan_id = 0; vlan_id < 4096; vlan_id++) {
if (rx_vft_set(port_id, vlan_id, on))
break;
}
}
void
vlan_tpid_set(portid_t port_id, enum rte_vlan_type vlan_type, uint16_t tp_id)
{
int diag;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
diag = rte_eth_dev_set_vlan_ether_type(port_id, vlan_type, tp_id);
if (diag == 0)
return;
printf("tx_vlan_tpid_set(port_pi=%d, vlan_type=%d, tpid=%d) failed "
"diag=%d\n",
port_id, vlan_type, tp_id, diag);
}
void
tx_vlan_set(portid_t port_id, uint16_t vlan_id)
{
int vlan_offload;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
if (vlan_id_is_invalid(vlan_id))
return;
vlan_offload = rte_eth_dev_get_vlan_offload(port_id);
if (vlan_offload & ETH_VLAN_EXTEND_OFFLOAD) {
printf("Error, as QinQ has been enabled.\n");
return;
}
tx_vlan_reset(port_id);
ports[port_id].tx_ol_flags |= TESTPMD_TX_OFFLOAD_INSERT_VLAN;
ports[port_id].tx_vlan_id = vlan_id;
}
void
tx_qinq_set(portid_t port_id, uint16_t vlan_id, uint16_t vlan_id_outer)
{
int vlan_offload;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
if (vlan_id_is_invalid(vlan_id))
return;
if (vlan_id_is_invalid(vlan_id_outer))
return;
vlan_offload = rte_eth_dev_get_vlan_offload(port_id);
if (!(vlan_offload & ETH_VLAN_EXTEND_OFFLOAD)) {
printf("Error, as QinQ hasn't been enabled.\n");
return;
}
tx_vlan_reset(port_id);
ports[port_id].tx_ol_flags |= TESTPMD_TX_OFFLOAD_INSERT_QINQ;
ports[port_id].tx_vlan_id = vlan_id;
ports[port_id].tx_vlan_id_outer = vlan_id_outer;
}
void
tx_vlan_reset(portid_t port_id)
{
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
ports[port_id].tx_ol_flags &= ~(TESTPMD_TX_OFFLOAD_INSERT_VLAN |
TESTPMD_TX_OFFLOAD_INSERT_QINQ);
ports[port_id].tx_vlan_id = 0;
ports[port_id].tx_vlan_id_outer = 0;
}
void
tx_vlan_pvid_set(portid_t port_id, uint16_t vlan_id, int on)
{
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
rte_eth_dev_set_vlan_pvid(port_id, vlan_id, on);
}
void
set_qmap(portid_t port_id, uint8_t is_rx, uint16_t queue_id, uint8_t map_value)
{
uint16_t i;
uint8_t existing_mapping_found = 0;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
if (is_rx ? (rx_queue_id_is_invalid(queue_id)) : (tx_queue_id_is_invalid(queue_id)))
return;
if (map_value >= RTE_ETHDEV_QUEUE_STAT_CNTRS) {
printf("map_value not in required range 0..%d\n",
RTE_ETHDEV_QUEUE_STAT_CNTRS - 1);
return;
}
if (!is_rx) { /*then tx*/
for (i = 0; i < nb_tx_queue_stats_mappings; i++) {
if ((tx_queue_stats_mappings[i].port_id == port_id) &&
(tx_queue_stats_mappings[i].queue_id == queue_id)) {
tx_queue_stats_mappings[i].stats_counter_id = map_value;
existing_mapping_found = 1;
break;
}
}
if (!existing_mapping_found) { /* A new additional mapping... */
tx_queue_stats_mappings[nb_tx_queue_stats_mappings].port_id = port_id;
tx_queue_stats_mappings[nb_tx_queue_stats_mappings].queue_id = queue_id;
tx_queue_stats_mappings[nb_tx_queue_stats_mappings].stats_counter_id = map_value;
nb_tx_queue_stats_mappings++;
}
}
else { /*rx*/
for (i = 0; i < nb_rx_queue_stats_mappings; i++) {
if ((rx_queue_stats_mappings[i].port_id == port_id) &&
(rx_queue_stats_mappings[i].queue_id == queue_id)) {
rx_queue_stats_mappings[i].stats_counter_id = map_value;
existing_mapping_found = 1;
break;
}
}
if (!existing_mapping_found) { /* A new additional mapping... */
rx_queue_stats_mappings[nb_rx_queue_stats_mappings].port_id = port_id;
rx_queue_stats_mappings[nb_rx_queue_stats_mappings].queue_id = queue_id;
rx_queue_stats_mappings[nb_rx_queue_stats_mappings].stats_counter_id = map_value;
nb_rx_queue_stats_mappings++;
}
}
}
static inline void
print_fdir_mask(struct rte_eth_fdir_masks *mask)
{
printf("\n vlan_tci: 0x%04x, ", mask->vlan_tci_mask);
if (fdir_conf.mode == RTE_FDIR_MODE_PERFECT_MAC_VLAN)
printf("mac_addr: 0x%02x", mask->mac_addr_byte_mask);
else if (fdir_conf.mode == RTE_FDIR_MODE_PERFECT_TUNNEL)
printf("mac_addr: 0x%02x, tunnel_type: 0x%01x, tunnel_id: 0x%08x",
mask->mac_addr_byte_mask, mask->tunnel_type_mask,
mask->tunnel_id_mask);
else {
printf("src_ipv4: 0x%08x, dst_ipv4: 0x%08x,"
" src_port: 0x%04x, dst_port: 0x%04x",
mask->ipv4_mask.src_ip, mask->ipv4_mask.dst_ip,
mask->src_port_mask, mask->dst_port_mask);
printf("\n src_ipv6: 0x%08x,0x%08x,0x%08x,0x%08x,"
" dst_ipv6: 0x%08x,0x%08x,0x%08x,0x%08x",
mask->ipv6_mask.src_ip[0], mask->ipv6_mask.src_ip[1],
mask->ipv6_mask.src_ip[2], mask->ipv6_mask.src_ip[3],
mask->ipv6_mask.dst_ip[0], mask->ipv6_mask.dst_ip[1],
mask->ipv6_mask.dst_ip[2], mask->ipv6_mask.dst_ip[3]);
}
printf("\n");
}
static inline void
print_fdir_flex_payload(struct rte_eth_fdir_flex_conf *flex_conf, uint32_t num)
{
struct rte_eth_flex_payload_cfg *cfg;
uint32_t i, j;
for (i = 0; i < flex_conf->nb_payloads; i++) {
cfg = &flex_conf->flex_set[i];
if (cfg->type == RTE_ETH_RAW_PAYLOAD)
printf("\n RAW: ");
else if (cfg->type == RTE_ETH_L2_PAYLOAD)
printf("\n L2_PAYLOAD: ");
else if (cfg->type == RTE_ETH_L3_PAYLOAD)
printf("\n L3_PAYLOAD: ");
else if (cfg->type == RTE_ETH_L4_PAYLOAD)
printf("\n L4_PAYLOAD: ");
else
printf("\n UNKNOWN PAYLOAD(%u): ", cfg->type);
for (j = 0; j < num; j++)
printf(" %-5u", cfg->src_offset[j]);
}
printf("\n");
}
static char *
flowtype_to_str(uint16_t flow_type)
{
struct flow_type_info {
char str[32];
uint16_t ftype;
};
uint8_t i;
static struct flow_type_info flowtype_str_table[] = {
{"raw", RTE_ETH_FLOW_RAW},
{"ipv4", RTE_ETH_FLOW_IPV4},
{"ipv4-frag", RTE_ETH_FLOW_FRAG_IPV4},
{"ipv4-tcp", RTE_ETH_FLOW_NONFRAG_IPV4_TCP},
{"ipv4-udp", RTE_ETH_FLOW_NONFRAG_IPV4_UDP},
{"ipv4-sctp", RTE_ETH_FLOW_NONFRAG_IPV4_SCTP},
{"ipv4-other", RTE_ETH_FLOW_NONFRAG_IPV4_OTHER},
{"ipv6", RTE_ETH_FLOW_IPV6},
{"ipv6-frag", RTE_ETH_FLOW_FRAG_IPV6},
{"ipv6-tcp", RTE_ETH_FLOW_NONFRAG_IPV6_TCP},
{"ipv6-udp", RTE_ETH_FLOW_NONFRAG_IPV6_UDP},
{"ipv6-sctp", RTE_ETH_FLOW_NONFRAG_IPV6_SCTP},
{"ipv6-other", RTE_ETH_FLOW_NONFRAG_IPV6_OTHER},
{"l2_payload", RTE_ETH_FLOW_L2_PAYLOAD},
};
for (i = 0; i < RTE_DIM(flowtype_str_table); i++) {
if (flowtype_str_table[i].ftype == flow_type)
return flowtype_str_table[i].str;
}
return NULL;
}
static inline void
print_fdir_flex_mask(struct rte_eth_fdir_flex_conf *flex_conf, uint32_t num)
{
struct rte_eth_fdir_flex_mask *mask;
uint32_t i, j;
char *p;
for (i = 0; i < flex_conf->nb_flexmasks; i++) {
mask = &flex_conf->flex_mask[i];
p = flowtype_to_str(mask->flow_type);
printf("\n %s:\t", p ? p : "unknown");
for (j = 0; j < num; j++)
printf(" %02x", mask->mask[j]);
}
printf("\n");
}
static inline void
print_fdir_flow_type(uint32_t flow_types_mask)
{
int i;
char *p;
for (i = RTE_ETH_FLOW_UNKNOWN; i < RTE_ETH_FLOW_MAX; i++) {
if (!(flow_types_mask & (1 << i)))
continue;
p = flowtype_to_str(i);
if (p)
printf(" %s", p);
else
printf(" unknown");
}
printf("\n");
}
void
fdir_get_infos(portid_t port_id)
{
struct rte_eth_fdir_stats fdir_stat;
struct rte_eth_fdir_info fdir_info;
int ret;
static const char *fdir_stats_border = "########################";
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
ret = rte_eth_dev_filter_supported(port_id, RTE_ETH_FILTER_FDIR);
if (ret < 0) {
printf("\n FDIR is not supported on port %-2d\n",
port_id);
return;
}
memset(&fdir_info, 0, sizeof(fdir_info));
rte_eth_dev_filter_ctrl(port_id, RTE_ETH_FILTER_FDIR,
RTE_ETH_FILTER_INFO, &fdir_info);
memset(&fdir_stat, 0, sizeof(fdir_stat));
rte_eth_dev_filter_ctrl(port_id, RTE_ETH_FILTER_FDIR,
RTE_ETH_FILTER_STATS, &fdir_stat);
printf("\n %s FDIR infos for port %-2d %s\n",
fdir_stats_border, port_id, fdir_stats_border);
printf(" MODE: ");
if (fdir_info.mode == RTE_FDIR_MODE_PERFECT)
printf(" PERFECT\n");
else if (fdir_info.mode == RTE_FDIR_MODE_PERFECT_MAC_VLAN)
printf(" PERFECT-MAC-VLAN\n");
else if (fdir_info.mode == RTE_FDIR_MODE_PERFECT_TUNNEL)
printf(" PERFECT-TUNNEL\n");
else if (fdir_info.mode == RTE_FDIR_MODE_SIGNATURE)
printf(" SIGNATURE\n");
else
printf(" DISABLE\n");
if (fdir_info.mode != RTE_FDIR_MODE_PERFECT_MAC_VLAN
&& fdir_info.mode != RTE_FDIR_MODE_PERFECT_TUNNEL) {
printf(" SUPPORTED FLOW TYPE: ");
print_fdir_flow_type(fdir_info.flow_types_mask[0]);
}
printf(" FLEX PAYLOAD INFO:\n");
printf(" max_len: %-10"PRIu32" payload_limit: %-10"PRIu32"\n"
" payload_unit: %-10"PRIu32" payload_seg: %-10"PRIu32"\n"
" bitmask_unit: %-10"PRIu32" bitmask_num: %-10"PRIu32"\n",
fdir_info.max_flexpayload, fdir_info.flex_payload_limit,
fdir_info.flex_payload_unit,
fdir_info.max_flex_payload_segment_num,
fdir_info.flex_bitmask_unit, fdir_info.max_flex_bitmask_num);
printf(" MASK: ");
print_fdir_mask(&fdir_info.mask);
if (fdir_info.flex_conf.nb_payloads > 0) {
printf(" FLEX PAYLOAD SRC OFFSET:");
print_fdir_flex_payload(&fdir_info.flex_conf, fdir_info.max_flexpayload);
}
if (fdir_info.flex_conf.nb_flexmasks > 0) {
printf(" FLEX MASK CFG:");
print_fdir_flex_mask(&fdir_info.flex_conf, fdir_info.max_flexpayload);
}
printf(" guarant_count: %-10"PRIu32" best_count: %"PRIu32"\n",
fdir_stat.guarant_cnt, fdir_stat.best_cnt);
printf(" guarant_space: %-10"PRIu32" best_space: %"PRIu32"\n",
fdir_info.guarant_spc, fdir_info.best_spc);
printf(" collision: %-10"PRIu32" free: %"PRIu32"\n"
" maxhash: %-10"PRIu32" maxlen: %"PRIu32"\n"
" add: %-10"PRIu64" remove: %"PRIu64"\n"
" f_add: %-10"PRIu64" f_remove: %"PRIu64"\n",
fdir_stat.collision, fdir_stat.free,
fdir_stat.maxhash, fdir_stat.maxlen,
fdir_stat.add, fdir_stat.remove,
fdir_stat.f_add, fdir_stat.f_remove);
printf(" %s############################%s\n",
fdir_stats_border, fdir_stats_border);
}
void
fdir_set_flex_mask(portid_t port_id, struct rte_eth_fdir_flex_mask *cfg)
{
struct rte_port *port;
struct rte_eth_fdir_flex_conf *flex_conf;
int i, idx = 0;
port = &ports[port_id];
flex_conf = &port->dev_conf.fdir_conf.flex_conf;
for (i = 0; i < RTE_ETH_FLOW_MAX; i++) {
if (cfg->flow_type == flex_conf->flex_mask[i].flow_type) {
idx = i;
break;
}
}
if (i >= RTE_ETH_FLOW_MAX) {
if (flex_conf->nb_flexmasks < RTE_DIM(flex_conf->flex_mask)) {
idx = flex_conf->nb_flexmasks;
flex_conf->nb_flexmasks++;
} else {
printf("The flex mask table is full. Can not set flex"
" mask for flow_type(%u).", cfg->flow_type);
return;
}
}
(void)rte_memcpy(&flex_conf->flex_mask[idx],
cfg,
sizeof(struct rte_eth_fdir_flex_mask));
}
void
fdir_set_flex_payload(portid_t port_id, struct rte_eth_flex_payload_cfg *cfg)
{
struct rte_port *port;
struct rte_eth_fdir_flex_conf *flex_conf;
int i, idx = 0;
port = &ports[port_id];
flex_conf = &port->dev_conf.fdir_conf.flex_conf;
for (i = 0; i < RTE_ETH_PAYLOAD_MAX; i++) {
if (cfg->type == flex_conf->flex_set[i].type) {
idx = i;
break;
}
}
if (i >= RTE_ETH_PAYLOAD_MAX) {
if (flex_conf->nb_payloads < RTE_DIM(flex_conf->flex_set)) {
idx = flex_conf->nb_payloads;
flex_conf->nb_payloads++;
} else {
printf("The flex payload table is full. Can not set"
" flex payload for type(%u).", cfg->type);
return;
}
}
(void)rte_memcpy(&flex_conf->flex_set[idx],
cfg,
sizeof(struct rte_eth_flex_payload_cfg));
}
void
set_vf_traffic(portid_t port_id, uint8_t is_rx, uint16_t vf, uint8_t on)
{
int diag;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
if (is_rx)
diag = rte_eth_dev_set_vf_rx(port_id,vf,on);
else
diag = rte_eth_dev_set_vf_tx(port_id,vf,on);
if (diag == 0)
return;
if(is_rx)
printf("rte_eth_dev_set_vf_rx for port_id=%d failed "
"diag=%d\n", port_id, diag);
else
printf("rte_eth_dev_set_vf_tx for port_id=%d failed "
"diag=%d\n", port_id, diag);
}
void
set_vf_rx_vlan(portid_t port_id, uint16_t vlan_id, uint64_t vf_mask, uint8_t on)
{
int diag;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
if (vlan_id_is_invalid(vlan_id))
return;
diag = rte_eth_dev_set_vf_vlan_filter(port_id, vlan_id, vf_mask, on);
if (diag == 0)
return;
printf("rte_eth_dev_set_vf_vlan_filter for port_id=%d failed "
"diag=%d\n", port_id, diag);
}
int
set_queue_rate_limit(portid_t port_id, uint16_t queue_idx, uint16_t rate)
{
int diag;
struct rte_eth_link link;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return 1;
rte_eth_link_get_nowait(port_id, &link);
if (rate > link.link_speed) {
printf("Invalid rate value:%u bigger than link speed: %u\n",
rate, link.link_speed);
return 1;
}
diag = rte_eth_set_queue_rate_limit(port_id, queue_idx, rate);
if (diag == 0)
return diag;
printf("rte_eth_set_queue_rate_limit for port_id=%d failed diag=%d\n",
port_id, diag);
return diag;
}
int
set_vf_rate_limit(portid_t port_id, uint16_t vf, uint16_t rate, uint64_t q_msk)
{
int diag;
struct rte_eth_link link;
if (q_msk == 0)
return 0;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return 1;
rte_eth_link_get_nowait(port_id, &link);
if (rate > link.link_speed) {
printf("Invalid rate value:%u bigger than link speed: %u\n",
rate, link.link_speed);
return 1;
}
diag = rte_eth_set_vf_rate_limit(port_id, vf, rate, q_msk);
if (diag == 0)
return diag;
printf("rte_eth_set_vf_rate_limit for port_id=%d failed diag=%d\n",
port_id, diag);
return diag;
}
/*
* Functions to manage the set of filtered Multicast MAC addresses.
*
* A pool of filtered multicast MAC addresses is associated with each port.
* The pool is allocated in chunks of MCAST_POOL_INC multicast addresses.
* The address of the pool and the number of valid multicast MAC addresses
* recorded in the pool are stored in the fields "mc_addr_pool" and
* "mc_addr_nb" of the "rte_port" data structure.
*
* The function "rte_eth_dev_set_mc_addr_list" of the PMDs API imposes
* to be supplied a contiguous array of multicast MAC addresses.
* To comply with this constraint, the set of multicast addresses recorded
* into the pool are systematically compacted at the beginning of the pool.
* Hence, when a multicast address is removed from the pool, all following
* addresses, if any, are copied back to keep the set contiguous.
*/
#define MCAST_POOL_INC 32
static int
mcast_addr_pool_extend(struct rte_port *port)
{
struct ether_addr *mc_pool;
size_t mc_pool_size;
/*
* If a free entry is available at the end of the pool, just
* increment the number of recorded multicast addresses.
*/
if ((port->mc_addr_nb % MCAST_POOL_INC) != 0) {
port->mc_addr_nb++;
return 0;
}
/*
* [re]allocate a pool with MCAST_POOL_INC more entries.
* The previous test guarantees that port->mc_addr_nb is a multiple
* of MCAST_POOL_INC.
*/
mc_pool_size = sizeof(struct ether_addr) * (port->mc_addr_nb +
MCAST_POOL_INC);
mc_pool = (struct ether_addr *) realloc(port->mc_addr_pool,
mc_pool_size);
if (mc_pool == NULL) {
printf("allocation of pool of %u multicast addresses failed\n",
port->mc_addr_nb + MCAST_POOL_INC);
return -ENOMEM;
}
port->mc_addr_pool = mc_pool;
port->mc_addr_nb++;
return 0;
}
static void
mcast_addr_pool_remove(struct rte_port *port, uint32_t addr_idx)
{
port->mc_addr_nb--;
if (addr_idx == port->mc_addr_nb) {
/* No need to recompact the set of multicast addressses. */
if (port->mc_addr_nb == 0) {
/* free the pool of multicast addresses. */
free(port->mc_addr_pool);
port->mc_addr_pool = NULL;
}
return;
}
memmove(&port->mc_addr_pool[addr_idx],
&port->mc_addr_pool[addr_idx + 1],
sizeof(struct ether_addr) * (port->mc_addr_nb - addr_idx));
}
static void
eth_port_multicast_addr_list_set(uint8_t port_id)
{
struct rte_port *port;
int diag;
port = &ports[port_id];
diag = rte_eth_dev_set_mc_addr_list(port_id, port->mc_addr_pool,
port->mc_addr_nb);
if (diag == 0)
return;
printf("rte_eth_dev_set_mc_addr_list(port=%d, nb=%u) failed. diag=%d\n",
port->mc_addr_nb, port_id, -diag);
}
void
mcast_addr_add(uint8_t port_id, struct ether_addr *mc_addr)
{
struct rte_port *port;
uint32_t i;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
port = &ports[port_id];
/*
* Check that the added multicast MAC address is not already recorded
* in the pool of multicast addresses.
*/
for (i = 0; i < port->mc_addr_nb; i++) {
if (is_same_ether_addr(mc_addr, &port->mc_addr_pool[i])) {
printf("multicast address already filtered by port\n");
return;
}
}
if (mcast_addr_pool_extend(port) != 0)
return;
ether_addr_copy(mc_addr, &port->mc_addr_pool[i]);
eth_port_multicast_addr_list_set(port_id);
}
void
mcast_addr_remove(uint8_t port_id, struct ether_addr *mc_addr)
{
struct rte_port *port;
uint32_t i;
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
port = &ports[port_id];
/*
* Search the pool of multicast MAC addresses for the removed address.
*/
for (i = 0; i < port->mc_addr_nb; i++) {
if (is_same_ether_addr(mc_addr, &port->mc_addr_pool[i]))
break;
}
if (i == port->mc_addr_nb) {
printf("multicast address not filtered by port %d\n", port_id);
return;
}
mcast_addr_pool_remove(port, i);
eth_port_multicast_addr_list_set(port_id);
}
void
port_dcb_info_display(uint8_t port_id)
{
struct rte_eth_dcb_info dcb_info;
uint16_t i;
int ret;
static const char *border = "================";
if (port_id_is_invalid(port_id, ENABLED_WARN))
return;
ret = rte_eth_dev_get_dcb_info(port_id, &dcb_info);
if (ret) {
printf("\n Failed to get dcb infos on port %-2d\n",
port_id);
return;
}
printf("\n %s DCB infos for port %-2d %s\n", border, port_id, border);
printf(" TC NUMBER: %d\n", dcb_info.nb_tcs);
printf("\n TC : ");
for (i = 0; i < dcb_info.nb_tcs; i++)
printf("\t%4d", i);
printf("\n Priority : ");
for (i = 0; i < dcb_info.nb_tcs; i++)
printf("\t%4d", dcb_info.prio_tc[i]);
printf("\n BW percent :");
for (i = 0; i < dcb_info.nb_tcs; i++)
printf("\t%4d%%", dcb_info.tc_bws[i]);
printf("\n RXQ base : ");
for (i = 0; i < dcb_info.nb_tcs; i++)
printf("\t%4d", dcb_info.tc_queue.tc_rxq[0][i].base);
printf("\n RXQ number :");
for (i = 0; i < dcb_info.nb_tcs; i++)
printf("\t%4d", dcb_info.tc_queue.tc_rxq[0][i].nb_queue);
printf("\n TXQ base : ");
for (i = 0; i < dcb_info.nb_tcs; i++)
printf("\t%4d", dcb_info.tc_queue.tc_txq[0][i].base);
printf("\n TXQ number :");
for (i = 0; i < dcb_info.nb_tcs; i++)
printf("\t%4d", dcb_info.tc_queue.tc_txq[0][i].nb_queue);
printf("\n");
}