-/*-
- * BSD LICENSE
- *
- * Copyright 2015 6WIND S.A.
- * Copyright 2015 Mellanox.
- *
- * 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.
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright 2015 6WIND S.A.
+ * Copyright 2015 Mellanox Technologies, Ltd
*/
-#define _GNU_SOURCE
-
#include <stddef.h>
-#include <assert.h>
#include <unistd.h>
-#include <stdint.h>
-#include <stdio.h>
#include <string.h>
+#include <stdint.h>
#include <stdlib.h>
#include <errno.h>
-#include <dirent.h>
-#include <net/if.h>
-#include <sys/ioctl.h>
-#include <sys/socket.h>
-#include <sys/utsname.h>
-#include <netinet/in.h>
-#include <linux/ethtool.h>
-#include <linux/sockios.h>
-#include <linux/version.h>
-#include <fcntl.h>
-#include <stdalign.h>
-#include <sys/un.h>
-#include <rte_atomic.h>
-#include <rte_ethdev.h>
+#include <ethdev_driver.h>
+#include <rte_bus_pci.h>
#include <rte_mbuf.h>
#include <rte_common.h>
#include <rte_interrupts.h>
-#include <rte_alarm.h>
#include <rte_malloc.h>
+#include <rte_string_fns.h>
+#include <rte_rwlock.h>
+#include <rte_cycles.h>
-#include "mlx5.h"
-#include "mlx5_rxtx.h"
-#include "mlx5_utils.h"
+#include <mlx5_malloc.h>
-/* Add defines in case the running kernel is not the same as user headers. */
-#ifndef ETHTOOL_GLINKSETTINGS
-struct ethtool_link_settings {
- uint32_t cmd;
- uint32_t speed;
- uint8_t duplex;
- uint8_t port;
- uint8_t phy_address;
- uint8_t autoneg;
- uint8_t mdio_support;
- uint8_t eth_to_mdix;
- uint8_t eth_tp_mdix_ctrl;
- int8_t link_mode_masks_nwords;
- uint32_t reserved[8];
- uint32_t link_mode_masks[];
-};
-
-#define ETHTOOL_GLINKSETTINGS 0x0000004c
-#define ETHTOOL_LINK_MODE_1000baseT_Full_BIT 5
-#define ETHTOOL_LINK_MODE_Autoneg_BIT 6
-#define ETHTOOL_LINK_MODE_1000baseKX_Full_BIT 17
-#define ETHTOOL_LINK_MODE_10000baseKX4_Full_BIT 18
-#define ETHTOOL_LINK_MODE_10000baseKR_Full_BIT 19
-#define ETHTOOL_LINK_MODE_10000baseR_FEC_BIT 20
-#define ETHTOOL_LINK_MODE_20000baseMLD2_Full_BIT 21
-#define ETHTOOL_LINK_MODE_20000baseKR2_Full_BIT 22
-#define ETHTOOL_LINK_MODE_40000baseKR4_Full_BIT 23
-#define ETHTOOL_LINK_MODE_40000baseCR4_Full_BIT 24
-#define ETHTOOL_LINK_MODE_40000baseSR4_Full_BIT 25
-#define ETHTOOL_LINK_MODE_40000baseLR4_Full_BIT 26
-#define ETHTOOL_LINK_MODE_56000baseKR4_Full_BIT 27
-#define ETHTOOL_LINK_MODE_56000baseCR4_Full_BIT 28
-#define ETHTOOL_LINK_MODE_56000baseSR4_Full_BIT 29
-#define ETHTOOL_LINK_MODE_56000baseLR4_Full_BIT 30
-#endif
-#ifndef HAVE_ETHTOOL_LINK_MODE_25G
-#define ETHTOOL_LINK_MODE_25000baseCR_Full_BIT 31
-#define ETHTOOL_LINK_MODE_25000baseKR_Full_BIT 32
-#define ETHTOOL_LINK_MODE_25000baseSR_Full_BIT 33
-#endif
-#ifndef HAVE_ETHTOOL_LINK_MODE_50G
-#define ETHTOOL_LINK_MODE_50000baseCR2_Full_BIT 34
-#define ETHTOOL_LINK_MODE_50000baseKR2_Full_BIT 35
-#endif
-#ifndef HAVE_ETHTOOL_LINK_MODE_100G
-#define ETHTOOL_LINK_MODE_100000baseKR4_Full_BIT 36
-#define ETHTOOL_LINK_MODE_100000baseSR4_Full_BIT 37
-#define ETHTOOL_LINK_MODE_100000baseCR4_Full_BIT 38
-#define ETHTOOL_LINK_MODE_100000baseLR4_ER4_Full_BIT 39
-#endif
+#include "mlx5_rxtx.h"
+#include "mlx5_rx.h"
+#include "mlx5_tx.h"
+#include "mlx5_autoconf.h"
/**
- * Return private structure associated with an Ethernet device.
+ * Get the interface index from device name.
*
- * @param dev
- * Pointer to Ethernet device structure.
+ * @param[in] dev
+ * Pointer to Ethernet device.
*
* @return
- * Pointer to private structure.
+ * Nonzero interface index on success, zero otherwise and rte_errno is set.
*/
-struct priv *
-mlx5_get_priv(struct rte_eth_dev *dev)
+unsigned int
+mlx5_ifindex(const struct rte_eth_dev *dev)
{
- return dev->data->dev_private;
-}
+ struct mlx5_priv *priv = dev->data->dev_private;
+ unsigned int ifindex;
-/**
- * Check if running as a secondary process.
- *
- * @return
- * Nonzero if running as a secondary process.
- */
-inline int
-mlx5_is_secondary(void)
-{
- return rte_eal_process_type() == RTE_PROC_SECONDARY;
+ MLX5_ASSERT(priv);
+ MLX5_ASSERT(priv->if_index);
+ if (priv->master && priv->sh->bond.ifindex > 0)
+ ifindex = priv->sh->bond.ifindex;
+ else
+ ifindex = priv->if_index;
+ if (!ifindex)
+ rte_errno = ENXIO;
+ return ifindex;
}
/**
- * Get interface name from private structure.
+ * DPDK callback for Ethernet device configuration.
*
- * @param[in] priv
- * Pointer to private structure.
- * @param[out] ifname
- * Interface name output buffer.
+ * @param dev
+ * Pointer to Ethernet device structure.
*
* @return
- * 0 on success, -1 on failure and errno is set.
+ * 0 on success, a negative errno value otherwise and rte_errno is set.
*/
int
-priv_get_ifname(const struct priv *priv, char (*ifname)[IF_NAMESIZE])
+mlx5_dev_configure(struct rte_eth_dev *dev)
{
- DIR *dir;
- struct dirent *dent;
- unsigned int dev_type = 0;
- unsigned int dev_port_prev = ~0u;
- char match[IF_NAMESIZE] = "";
-
- {
- MKSTR(path, "%s/device/net", priv->ibdev_path);
+ struct mlx5_priv *priv = dev->data->dev_private;
+ unsigned int rxqs_n = dev->data->nb_rx_queues;
+ unsigned int txqs_n = dev->data->nb_tx_queues;
+ const uint8_t use_app_rss_key =
+ !!dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key;
+ int ret = 0;
- dir = opendir(path);
- if (dir == NULL)
- return -1;
+ if (use_app_rss_key &&
+ (dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key_len !=
+ MLX5_RSS_HASH_KEY_LEN)) {
+ DRV_LOG(ERR, "port %u RSS key len must be %s Bytes long",
+ dev->data->port_id, RTE_STR(MLX5_RSS_HASH_KEY_LEN));
+ rte_errno = EINVAL;
+ return -rte_errno;
}
- while ((dent = readdir(dir)) != NULL) {
- char *name = dent->d_name;
- FILE *file;
- unsigned int dev_port;
- int r;
-
- if ((name[0] == '.') &&
- ((name[1] == '\0') ||
- ((name[1] == '.') && (name[2] == '\0'))))
- continue;
-
- MKSTR(path, "%s/device/net/%s/%s",
- priv->ibdev_path, name,
- (dev_type ? "dev_id" : "dev_port"));
-
- file = fopen(path, "rb");
- if (file == NULL) {
- if (errno != ENOENT)
- continue;
- /*
- * Switch to dev_id when dev_port does not exist as
- * is the case with Linux kernel versions < 3.15.
- */
-try_dev_id:
- match[0] = '\0';
- if (dev_type)
- break;
- dev_type = 1;
- dev_port_prev = ~0u;
- rewinddir(dir);
- continue;
- }
- r = fscanf(file, (dev_type ? "%x" : "%u"), &dev_port);
- fclose(file);
- if (r != 1)
- continue;
- /*
- * Switch to dev_id when dev_port returns the same value for
- * all ports. May happen when using a MOFED release older than
- * 3.0 with a Linux kernel >= 3.15.
- */
- if (dev_port == dev_port_prev)
- goto try_dev_id;
- dev_port_prev = dev_port;
- if (dev_port == (priv->port - 1u))
- snprintf(match, sizeof(match), "%s", name);
+ priv->rss_conf.rss_key =
+ mlx5_realloc(priv->rss_conf.rss_key, MLX5_MEM_RTE,
+ MLX5_RSS_HASH_KEY_LEN, 0, SOCKET_ID_ANY);
+ if (!priv->rss_conf.rss_key) {
+ DRV_LOG(ERR, "port %u cannot allocate RSS hash key memory (%u)",
+ dev->data->port_id, rxqs_n);
+ rte_errno = ENOMEM;
+ return -rte_errno;
}
- closedir(dir);
- if (match[0] == '\0')
- return -1;
- strncpy(*ifname, match, sizeof(*ifname));
- return 0;
-}
-
-/**
- * Check if the counter is located on ib counters file.
- *
- * @param[in] cntr
- * Counter name.
- *
- * @return
- * 1 if counter is located on ib counters file , 0 otherwise.
- */
-int
-priv_is_ib_cntr(const char *cntr)
-{
- if (!strcmp(cntr, "out_of_buffer"))
- return 1;
- return 0;
-}
-
-/**
- * Read from sysfs entry.
- *
- * @param[in] priv
- * Pointer to private structure.
- * @param[in] entry
- * Entry name relative to sysfs path.
- * @param[out] buf
- * Data output buffer.
- * @param size
- * Buffer size.
- *
- * @return
- * 0 on success, -1 on failure and errno is set.
- */
-static int
-priv_sysfs_read(const struct priv *priv, const char *entry,
- char *buf, size_t size)
-{
- char ifname[IF_NAMESIZE];
- FILE *file;
- int ret;
- int err;
- if (priv_get_ifname(priv, &ifname))
- return -1;
-
- if (priv_is_ib_cntr(entry)) {
- MKSTR(path, "%s/ports/1/hw_counters/%s",
- priv->ibdev_path, entry);
- file = fopen(path, "rb");
- } else {
- MKSTR(path, "%s/device/net/%s/%s",
- priv->ibdev_path, ifname, entry);
- file = fopen(path, "rb");
+ if ((dev->data->dev_conf.txmode.offloads &
+ DEV_TX_OFFLOAD_SEND_ON_TIMESTAMP) &&
+ rte_mbuf_dyn_tx_timestamp_register(NULL, NULL) != 0) {
+ DRV_LOG(ERR, "port %u cannot register Tx timestamp field/flag",
+ dev->data->port_id);
+ return -rte_errno;
}
- if (file == NULL)
- return -1;
- ret = fread(buf, 1, size, file);
- err = errno;
- if (((size_t)ret < size) && (ferror(file)))
- ret = -1;
- else
- ret = size;
- fclose(file);
- errno = err;
- return ret;
-}
-
-/**
- * Write to sysfs entry.
- *
- * @param[in] priv
- * Pointer to private structure.
- * @param[in] entry
- * Entry name relative to sysfs path.
- * @param[in] buf
- * Data buffer.
- * @param size
- * Buffer size.
- *
- * @return
- * 0 on success, -1 on failure and errno is set.
- */
-static int
-priv_sysfs_write(const struct priv *priv, const char *entry,
- char *buf, size_t size)
-{
- char ifname[IF_NAMESIZE];
- FILE *file;
- int ret;
- int err;
-
- if (priv_get_ifname(priv, &ifname))
- return -1;
-
- MKSTR(path, "%s/device/net/%s/%s", priv->ibdev_path, ifname, entry);
-
- file = fopen(path, "wb");
- if (file == NULL)
- return -1;
- ret = fwrite(buf, 1, size, file);
- err = errno;
- if (((size_t)ret < size) || (ferror(file)))
- ret = -1;
- else
- ret = size;
- fclose(file);
- errno = err;
- return ret;
-}
-
-/**
- * Get unsigned long sysfs property.
- *
- * @param priv
- * Pointer to private structure.
- * @param[in] name
- * Entry name relative to sysfs path.
- * @param[out] value
- * Value output buffer.
- *
- * @return
- * 0 on success, -1 on failure and errno is set.
- */
-static int
-priv_get_sysfs_ulong(struct priv *priv, const char *name, unsigned long *value)
-{
- int ret;
- unsigned long value_ret;
- char value_str[32];
-
- ret = priv_sysfs_read(priv, name, value_str, (sizeof(value_str) - 1));
- if (ret == -1) {
- DEBUG("cannot read %s value from sysfs: %s",
- name, strerror(errno));
- return -1;
+ memcpy(priv->rss_conf.rss_key,
+ use_app_rss_key ?
+ dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key :
+ rss_hash_default_key,
+ MLX5_RSS_HASH_KEY_LEN);
+ priv->rss_conf.rss_key_len = MLX5_RSS_HASH_KEY_LEN;
+ priv->rss_conf.rss_hf = dev->data->dev_conf.rx_adv_conf.rss_conf.rss_hf;
+ priv->rxqs = (void *)dev->data->rx_queues;
+ priv->txqs = (void *)dev->data->tx_queues;
+ if (txqs_n != priv->txqs_n) {
+ DRV_LOG(INFO, "port %u Tx queues number update: %u -> %u",
+ dev->data->port_id, priv->txqs_n, txqs_n);
+ priv->txqs_n = txqs_n;
}
- value_str[ret] = '\0';
- errno = 0;
- value_ret = strtoul(value_str, NULL, 0);
- if (errno) {
- DEBUG("invalid %s value `%s': %s", name, value_str,
- strerror(errno));
- return -1;
+ if (rxqs_n > priv->config.ind_table_max_size) {
+ DRV_LOG(ERR, "port %u cannot handle this many Rx queues (%u)",
+ dev->data->port_id, rxqs_n);
+ rte_errno = EINVAL;
+ return -rte_errno;
}
- *value = value_ret;
- return 0;
-}
-
-/**
- * Set unsigned long sysfs property.
- *
- * @param priv
- * Pointer to private structure.
- * @param[in] name
- * Entry name relative to sysfs path.
- * @param value
- * Value to set.
- *
- * @return
- * 0 on success, -1 on failure and errno is set.
- */
-static int
-priv_set_sysfs_ulong(struct priv *priv, const char *name, unsigned long value)
-{
- int ret;
- MKSTR(value_str, "%lu", value);
-
- ret = priv_sysfs_write(priv, name, value_str, (sizeof(value_str) - 1));
- if (ret == -1) {
- DEBUG("cannot write %s `%s' (%lu) to sysfs: %s",
- name, value_str, value, strerror(errno));
- return -1;
+ if (rxqs_n != priv->rxqs_n) {
+ DRV_LOG(INFO, "port %u Rx queues number update: %u -> %u",
+ dev->data->port_id, priv->rxqs_n, rxqs_n);
+ priv->rxqs_n = rxqs_n;
}
+ priv->skip_default_rss_reta = 0;
+ ret = mlx5_proc_priv_init(dev);
+ if (ret)
+ return ret;
return 0;
}
/**
- * Perform ifreq ioctl() on associated Ethernet device.
+ * Configure default RSS reta.
*
- * @param[in] priv
- * Pointer to private structure.
- * @param req
- * Request number to pass to ioctl().
- * @param[out] ifr
- * Interface request structure output buffer.
+ * @param dev
+ * Pointer to Ethernet device structure.
*
* @return
- * 0 on success, -1 on failure and errno is set.
+ * 0 on success, a negative errno value otherwise and rte_errno is set.
*/
int
-priv_ifreq(const struct priv *priv, int req, struct ifreq *ifr)
+mlx5_dev_configure_rss_reta(struct rte_eth_dev *dev)
{
- int sock = socket(PF_INET, SOCK_DGRAM, IPPROTO_IP);
- int ret = -1;
+ struct mlx5_priv *priv = dev->data->dev_private;
+ unsigned int rxqs_n = dev->data->nb_rx_queues;
+ unsigned int i;
+ unsigned int j;
+ unsigned int reta_idx_n;
+ int ret = 0;
+ unsigned int *rss_queue_arr = NULL;
+ unsigned int rss_queue_n = 0;
- if (sock == -1)
+ if (priv->skip_default_rss_reta)
return ret;
- if (priv_get_ifname(priv, &ifr->ifr_name) == 0)
- ret = ioctl(sock, req, ifr);
- close(sock);
+ rss_queue_arr = mlx5_malloc(0, rxqs_n * sizeof(unsigned int), 0,
+ SOCKET_ID_ANY);
+ if (!rss_queue_arr) {
+ DRV_LOG(ERR, "port %u cannot allocate RSS queue list (%u)",
+ dev->data->port_id, rxqs_n);
+ rte_errno = ENOMEM;
+ return -rte_errno;
+ }
+ for (i = 0, j = 0; i < rxqs_n; i++) {
+ struct mlx5_rxq_data *rxq_data;
+ struct mlx5_rxq_ctrl *rxq_ctrl;
+
+ rxq_data = (*priv->rxqs)[i];
+ rxq_ctrl = container_of(rxq_data, struct mlx5_rxq_ctrl, rxq);
+ if (rxq_ctrl && rxq_ctrl->type == MLX5_RXQ_TYPE_STANDARD)
+ rss_queue_arr[j++] = i;
+ }
+ rss_queue_n = j;
+ if (rss_queue_n > priv->config.ind_table_max_size) {
+ DRV_LOG(ERR, "port %u cannot handle this many Rx queues (%u)",
+ dev->data->port_id, rss_queue_n);
+ rte_errno = EINVAL;
+ mlx5_free(rss_queue_arr);
+ return -rte_errno;
+ }
+ DRV_LOG(INFO, "port %u Rx queues number update: %u -> %u",
+ dev->data->port_id, priv->rxqs_n, rxqs_n);
+ priv->rxqs_n = rxqs_n;
+ /*
+ * If the requested number of RX queues is not a power of two,
+ * use the maximum indirection table size for better balancing.
+ * The result is always rounded to the next power of two.
+ */
+ reta_idx_n = (1 << log2above((rss_queue_n & (rss_queue_n - 1)) ?
+ priv->config.ind_table_max_size :
+ rss_queue_n));
+ ret = mlx5_rss_reta_index_resize(dev, reta_idx_n);
+ if (ret) {
+ mlx5_free(rss_queue_arr);
+ return ret;
+ }
+ /*
+ * When the number of RX queues is not a power of two,
+ * the remaining table entries are padded with reused WQs
+ * and hashes are not spread uniformly.
+ */
+ for (i = 0, j = 0; (i != reta_idx_n); ++i) {
+ (*priv->reta_idx)[i] = rss_queue_arr[j];
+ if (++j == rss_queue_n)
+ j = 0;
+ }
+ mlx5_free(rss_queue_arr);
return ret;
}
/**
- * Return the number of active VFs for the current device.
+ * Sets default tuning parameters.
*
- * @param[in] priv
- * Pointer to private structure.
- * @param[out] num_vfs
- * Number of active VFs.
- *
- * @return
- * 0 on success, -1 on failure and errno is set.
+ * @param dev
+ * Pointer to Ethernet device.
+ * @param[out] info
+ * Info structure output buffer.
*/
-int
-priv_get_num_vfs(struct priv *priv, uint16_t *num_vfs)
-{
- /* The sysfs entry name depends on the operating system. */
- const char **name = (const char *[]){
- "device/sriov_numvfs",
- "device/mlx5_num_vfs",
- NULL,
- };
- int ret;
-
- do {
- unsigned long ulong_num_vfs;
-
- ret = priv_get_sysfs_ulong(priv, *name, &ulong_num_vfs);
- if (!ret)
- *num_vfs = ulong_num_vfs;
- } while (*(++name) && ret);
- return ret;
+static void
+mlx5_set_default_params(struct rte_eth_dev *dev, struct rte_eth_dev_info *info)
+{
+ struct mlx5_priv *priv = dev->data->dev_private;
+
+ /* Minimum CPU utilization. */
+ info->default_rxportconf.ring_size = 256;
+ info->default_txportconf.ring_size = 256;
+ info->default_rxportconf.burst_size = MLX5_RX_DEFAULT_BURST;
+ info->default_txportconf.burst_size = MLX5_TX_DEFAULT_BURST;
+ if ((priv->link_speed_capa & ETH_LINK_SPEED_200G) |
+ (priv->link_speed_capa & ETH_LINK_SPEED_100G)) {
+ info->default_rxportconf.nb_queues = 16;
+ info->default_txportconf.nb_queues = 16;
+ if (dev->data->nb_rx_queues > 2 ||
+ dev->data->nb_tx_queues > 2) {
+ /* Max Throughput. */
+ info->default_rxportconf.ring_size = 2048;
+ info->default_txportconf.ring_size = 2048;
+ }
+ } else {
+ info->default_rxportconf.nb_queues = 8;
+ info->default_txportconf.nb_queues = 8;
+ if (dev->data->nb_rx_queues > 2 ||
+ dev->data->nb_tx_queues > 2) {
+ /* Max Throughput. */
+ info->default_rxportconf.ring_size = 4096;
+ info->default_txportconf.ring_size = 4096;
+ }
+ }
}
/**
- * Get device MTU.
- *
- * @param priv
- * Pointer to private structure.
- * @param[out] mtu
- * MTU value output buffer.
+ * Sets tx mbuf limiting parameters.
*
- * @return
- * 0 on success, -1 on failure and errno is set.
+ * @param dev
+ * Pointer to Ethernet device.
+ * @param[out] info
+ * Info structure output buffer.
*/
-int
-priv_get_mtu(struct priv *priv, uint16_t *mtu)
-{
- unsigned long ulong_mtu;
-
- if (priv_get_sysfs_ulong(priv, "mtu", &ulong_mtu) == -1)
- return -1;
- *mtu = ulong_mtu;
- return 0;
+static void
+mlx5_set_txlimit_params(struct rte_eth_dev *dev, struct rte_eth_dev_info *info)
+{
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct mlx5_dev_config *config = &priv->config;
+ unsigned int inlen;
+ uint16_t nb_max;
+
+ inlen = (config->txq_inline_max == MLX5_ARG_UNSET) ?
+ MLX5_SEND_DEF_INLINE_LEN :
+ (unsigned int)config->txq_inline_max;
+ MLX5_ASSERT(config->txq_inline_min >= 0);
+ inlen = RTE_MAX(inlen, (unsigned int)config->txq_inline_min);
+ inlen = RTE_MIN(inlen, MLX5_WQE_SIZE_MAX +
+ MLX5_ESEG_MIN_INLINE_SIZE -
+ MLX5_WQE_CSEG_SIZE -
+ MLX5_WQE_ESEG_SIZE -
+ MLX5_WQE_DSEG_SIZE * 2);
+ nb_max = (MLX5_WQE_SIZE_MAX +
+ MLX5_ESEG_MIN_INLINE_SIZE -
+ MLX5_WQE_CSEG_SIZE -
+ MLX5_WQE_ESEG_SIZE -
+ MLX5_WQE_DSEG_SIZE -
+ inlen) / MLX5_WSEG_SIZE;
+ info->tx_desc_lim.nb_seg_max = nb_max;
+ info->tx_desc_lim.nb_mtu_seg_max = nb_max;
}
/**
- * Read device counter from sysfs.
- *
- * @param priv
- * Pointer to private structure.
- * @param name
- * Counter name.
- * @param[out] cntr
- * Counter output buffer.
+ * DPDK callback to get information about the device.
*
- * @return
- * 0 on success, -1 on failure and errno is set.
+ * @param dev
+ * Pointer to Ethernet device structure.
+ * @param[out] info
+ * Info structure output buffer.
*/
int
-priv_get_cntr_sysfs(struct priv *priv, const char *name, uint64_t *cntr)
+mlx5_dev_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *info)
{
- unsigned long ulong_ctr;
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct mlx5_dev_config *config = &priv->config;
+ unsigned int max;
- if (priv_get_sysfs_ulong(priv, name, &ulong_ctr) == -1)
- return -1;
- *cntr = ulong_ctr;
+ /* FIXME: we should ask the device for these values. */
+ info->min_rx_bufsize = 32;
+ info->max_rx_pktlen = 65536;
+ info->max_lro_pkt_size = MLX5_MAX_LRO_SIZE;
+ /*
+ * Since we need one CQ per QP, the limit is the minimum number
+ * between the two values.
+ */
+ max = RTE_MIN(priv->sh->device_attr.max_cq,
+ priv->sh->device_attr.max_qp);
+ /* max_rx_queues is uint16_t. */
+ max = RTE_MIN(max, (unsigned int)UINT16_MAX);
+ info->max_rx_queues = max;
+ info->max_tx_queues = max;
+ info->max_mac_addrs = MLX5_MAX_UC_MAC_ADDRESSES;
+ info->rx_queue_offload_capa = mlx5_get_rx_queue_offloads(dev);
+ info->rx_seg_capa.max_nseg = MLX5_MAX_RXQ_NSEG;
+ info->rx_seg_capa.multi_pools = !config->mprq.enabled;
+ info->rx_seg_capa.offset_allowed = !config->mprq.enabled;
+ info->rx_seg_capa.offset_align_log2 = 0;
+ info->rx_offload_capa = (mlx5_get_rx_port_offloads() |
+ info->rx_queue_offload_capa);
+ info->tx_offload_capa = mlx5_get_tx_port_offloads(dev);
+ info->if_index = mlx5_ifindex(dev);
+ info->reta_size = priv->reta_idx_n ?
+ priv->reta_idx_n : config->ind_table_max_size;
+ info->hash_key_size = MLX5_RSS_HASH_KEY_LEN;
+ info->speed_capa = priv->link_speed_capa;
+ info->flow_type_rss_offloads = ~MLX5_RSS_HF_MASK;
+ mlx5_set_default_params(dev, info);
+ mlx5_set_txlimit_params(dev, info);
+ info->switch_info.name = dev->data->name;
+ info->switch_info.domain_id = priv->domain_id;
+ info->switch_info.port_id = priv->representor_id;
+ if (priv->representor) {
+ uint16_t port_id;
+
+ MLX5_ETH_FOREACH_DEV(port_id, priv->pci_dev) {
+ struct mlx5_priv *opriv =
+ rte_eth_devices[port_id].data->dev_private;
+
+ if (!opriv ||
+ opriv->representor ||
+ opriv->sh != priv->sh ||
+ opriv->domain_id != priv->domain_id)
+ continue;
+ /*
+ * Override switch name with that of the master
+ * device.
+ */
+ info->switch_info.name = opriv->dev_data->name;
+ break;
+ }
+ }
return 0;
}
/**
- * Set device MTU.
+ * Calculate representor ID from port switch info.
*
- * @param priv
- * Pointer to private structure.
- * @param mtu
- * MTU value to set.
+ * Uint16 representor ID bits definition:
+ * pf: 2
+ * type: 2
+ * vf/sf: 12
*
- * @return
- * 0 on success, -1 on failure and errno is set.
- */
-static int
-priv_set_mtu(struct priv *priv, uint16_t mtu)
-{
- uint16_t new_mtu;
-
- if (priv_set_sysfs_ulong(priv, "mtu", mtu) ||
- priv_get_mtu(priv, &new_mtu))
- return -1;
- if (new_mtu == mtu)
- return 0;
- errno = EINVAL;
- return -1;
-}
-
-/**
- * Set device flags.
- *
- * @param priv
- * Pointer to private structure.
- * @param keep
- * Bitmask for flags that must remain untouched.
- * @param flags
- * Bitmask for flags to modify.
+ * @param info
+ * Port switch info.
+ * @param hpf_type
+ * Use this type if port is HPF.
*
* @return
- * 0 on success, -1 on failure and errno is set.
- */
-int
-priv_set_flags(struct priv *priv, unsigned int keep, unsigned int flags)
-{
- unsigned long tmp;
-
- if (priv_get_sysfs_ulong(priv, "flags", &tmp) == -1)
- return -1;
- tmp &= keep;
- tmp |= (flags & (~keep));
- return priv_set_sysfs_ulong(priv, "flags", tmp);
+ * Encoded representor ID.
+ */
+uint16_t
+mlx5_representor_id_encode(const struct mlx5_switch_info *info,
+ enum rte_eth_representor_type hpf_type)
+{
+ enum rte_eth_representor_type type = RTE_ETH_REPRESENTOR_VF;
+ uint16_t repr = info->port_name;
+
+ if (info->representor == 0)
+ return UINT16_MAX;
+ if (info->name_type == MLX5_PHYS_PORT_NAME_TYPE_PFSF)
+ type = RTE_ETH_REPRESENTOR_SF;
+ if (info->name_type == MLX5_PHYS_PORT_NAME_TYPE_PFHPF) {
+ type = hpf_type;
+ repr = UINT16_MAX;
+ }
+ return MLX5_REPRESENTOR_ID(info->pf_num, type, repr);
}
/**
- * Ethernet device configuration.
+ * DPDK callback to get information about representor.
*
- * Prepare the driver for a given number of TX and RX queues.
+ * Representor ID bits definition:
+ * vf/sf: 12
+ * type: 2
+ * pf: 2
*
* @param dev
* Pointer to Ethernet device structure.
+ * @param[out] info
+ * Nullable info structure output buffer.
*
* @return
- * 0 on success, errno value on failure.
+ * negative on error, or the number of representor ranges.
*/
-static int
-dev_configure(struct rte_eth_dev *dev)
+int
+mlx5_representor_info_get(struct rte_eth_dev *dev,
+ struct rte_eth_representor_info *info)
{
- struct priv *priv = dev->data->dev_private;
- unsigned int rxqs_n = dev->data->nb_rx_queues;
- unsigned int txqs_n = dev->data->nb_tx_queues;
- unsigned int i;
- unsigned int j;
- unsigned int reta_idx_n;
- const uint8_t use_app_rss_key =
- !!dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key_len;
+ struct mlx5_priv *priv = dev->data->dev_private;
+ int n_type = 4; /* Representor types, VF, HPF@VF, SF and HPF@SF. */
+ int n_pf = 2; /* Number of PFs. */
+ int i = 0, pf;
- if (use_app_rss_key &&
- (dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key_len !=
- rss_hash_default_key_len)) {
- /* MLX5 RSS only support 40bytes key. */
- return EINVAL;
- }
- priv->rss_conf.rss_key =
- rte_realloc(priv->rss_conf.rss_key,
- rss_hash_default_key_len, 0);
- if (!priv->rss_conf.rss_key) {
- ERROR("cannot allocate RSS hash key memory (%u)", rxqs_n);
- return ENOMEM;
- }
- memcpy(priv->rss_conf.rss_key,
- use_app_rss_key ?
- dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key :
- rss_hash_default_key,
- rss_hash_default_key_len);
- priv->rss_conf.rss_key_len = rss_hash_default_key_len;
- priv->rss_conf.rss_hf = dev->data->dev_conf.rx_adv_conf.rss_conf.rss_hf;
- priv->rxqs = (void *)dev->data->rx_queues;
- priv->txqs = (void *)dev->data->tx_queues;
- if (txqs_n != priv->txqs_n) {
- INFO("%p: TX queues number update: %u -> %u",
- (void *)dev, priv->txqs_n, txqs_n);
- priv->txqs_n = txqs_n;
- }
- if (rxqs_n > priv->ind_table_max_size) {
- ERROR("cannot handle this many RX queues (%u)", rxqs_n);
- return EINVAL;
- }
- if (rxqs_n == priv->rxqs_n)
- return 0;
- INFO("%p: RX queues number update: %u -> %u",
- (void *)dev, priv->rxqs_n, rxqs_n);
- priv->rxqs_n = rxqs_n;
- /* If the requested number of RX queues is not a power of two, use the
- * maximum indirection table size for better balancing.
- * The result is always rounded to the next power of two. */
- reta_idx_n = (1 << log2above((rxqs_n & (rxqs_n - 1)) ?
- priv->ind_table_max_size :
- rxqs_n));
- if (priv_rss_reta_index_resize(priv, reta_idx_n))
- return ENOMEM;
- /* When the number of RX queues is not a power of two, the remaining
- * table entries are padded with reused WQs and hashes are not spread
- * uniformly. */
- for (i = 0, j = 0; (i != reta_idx_n); ++i) {
- (*priv->reta_idx)[i] = j;
- if (++j == rxqs_n)
- j = 0;
+ if (info == NULL)
+ goto out;
+ info->controller = 0;
+ info->pf = priv->pf_bond >= 0 ? priv->pf_bond : 0;
+ for (pf = 0; pf < n_pf; ++pf) {
+ /* VF range. */
+ info->ranges[i].type = RTE_ETH_REPRESENTOR_VF;
+ info->ranges[i].controller = 0;
+ info->ranges[i].pf = pf;
+ info->ranges[i].vf = 0;
+ info->ranges[i].id_base =
+ MLX5_REPRESENTOR_ID(pf, info->ranges[i].type, 0);
+ info->ranges[i].id_end =
+ MLX5_REPRESENTOR_ID(pf, info->ranges[i].type, -1);
+ snprintf(info->ranges[i].name,
+ sizeof(info->ranges[i].name), "pf%dvf", pf);
+ i++;
+ /* HPF range of VF type. */
+ info->ranges[i].type = RTE_ETH_REPRESENTOR_VF;
+ info->ranges[i].controller = 0;
+ info->ranges[i].pf = pf;
+ info->ranges[i].vf = UINT16_MAX;
+ info->ranges[i].id_base =
+ MLX5_REPRESENTOR_ID(pf, info->ranges[i].type, -1);
+ info->ranges[i].id_end =
+ MLX5_REPRESENTOR_ID(pf, info->ranges[i].type, -1);
+ snprintf(info->ranges[i].name,
+ sizeof(info->ranges[i].name), "pf%dvf", pf);
+ i++;
+ /* SF range. */
+ info->ranges[i].type = RTE_ETH_REPRESENTOR_SF;
+ info->ranges[i].controller = 0;
+ info->ranges[i].pf = pf;
+ info->ranges[i].vf = 0;
+ info->ranges[i].id_base =
+ MLX5_REPRESENTOR_ID(pf, info->ranges[i].type, 0);
+ info->ranges[i].id_end =
+ MLX5_REPRESENTOR_ID(pf, info->ranges[i].type, -1);
+ snprintf(info->ranges[i].name,
+ sizeof(info->ranges[i].name), "pf%dsf", pf);
+ i++;
+ /* HPF range of SF type. */
+ info->ranges[i].type = RTE_ETH_REPRESENTOR_SF;
+ info->ranges[i].controller = 0;
+ info->ranges[i].pf = pf;
+ info->ranges[i].vf = UINT16_MAX;
+ info->ranges[i].id_base =
+ MLX5_REPRESENTOR_ID(pf, info->ranges[i].type, -1);
+ info->ranges[i].id_end =
+ MLX5_REPRESENTOR_ID(pf, info->ranges[i].type, -1);
+ snprintf(info->ranges[i].name,
+ sizeof(info->ranges[i].name), "pf%dsf", pf);
+ i++;
}
- return 0;
+out:
+ return n_type * n_pf;
}
/**
- * DPDK callback for Ethernet device configuration.
+ * Get firmware version of a device.
*
* @param dev
- * Pointer to Ethernet device structure.
+ * Ethernet device port.
+ * @param fw_ver
+ * String output allocated by caller.
+ * @param fw_size
+ * Size of the output string, including terminating null byte.
*
* @return
- * 0 on success, negative errno value on failure.
+ * 0 on success, or the size of the non truncated string if too big.
*/
int
-mlx5_dev_configure(struct rte_eth_dev *dev)
+mlx5_fw_version_get(struct rte_eth_dev *dev, char *fw_ver, size_t fw_size)
{
- struct priv *priv = dev->data->dev_private;
- int ret;
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct mlx5_dev_attr *attr = &priv->sh->device_attr;
+ size_t size = strnlen(attr->fw_ver, sizeof(attr->fw_ver)) + 1;
- if (mlx5_is_secondary())
- return -E_RTE_SECONDARY;
-
- priv_lock(priv);
- ret = dev_configure(dev);
- assert(ret >= 0);
- priv_unlock(priv);
- return -ret;
+ if (fw_size < size)
+ return size;
+ if (fw_ver != NULL)
+ strlcpy(fw_ver, attr->fw_ver, fw_size);
+ return 0;
}
/**
- * DPDK callback to get information about the device.
+ * Get supported packet types.
*
* @param dev
* Pointer to Ethernet device structure.
- * @param[out] info
- * Info structure output buffer.
+ *
+ * @return
+ * A pointer to the supported Packet types array.
*/
-void
-mlx5_dev_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *info)
-{
- struct priv *priv = mlx5_get_priv(dev);
- unsigned int max;
- char ifname[IF_NAMESIZE];
-
- info->pci_dev = RTE_ETH_DEV_TO_PCI(dev);
-
- priv_lock(priv);
- /* FIXME: we should ask the device for these values. */
- info->min_rx_bufsize = 32;
- info->max_rx_pktlen = 65536;
- /*
- * Since we need one CQ per QP, the limit is the minimum number
- * between the two values.
- */
- max = RTE_MIN(priv->device_attr.orig_attr.max_cq,
- priv->device_attr.orig_attr.max_qp);
- /* If max >= 65535 then max = 0, max_rx_queues is uint16_t. */
- if (max >= 65535)
- max = 65535;
- info->max_rx_queues = max;
- info->max_tx_queues = max;
- info->max_mac_addrs = RTE_DIM(priv->mac);
- info->rx_offload_capa =
- (priv->hw_csum ?
- (DEV_RX_OFFLOAD_IPV4_CKSUM |
- DEV_RX_OFFLOAD_UDP_CKSUM |
- DEV_RX_OFFLOAD_TCP_CKSUM) :
- 0) |
- (priv->hw_vlan_strip ? DEV_RX_OFFLOAD_VLAN_STRIP : 0);
- if (!priv->mps)
- info->tx_offload_capa = DEV_TX_OFFLOAD_VLAN_INSERT;
- if (priv->hw_csum)
- info->tx_offload_capa |=
- (DEV_TX_OFFLOAD_IPV4_CKSUM |
- DEV_TX_OFFLOAD_UDP_CKSUM |
- DEV_TX_OFFLOAD_TCP_CKSUM);
- if (priv->tso)
- info->tx_offload_capa |= DEV_TX_OFFLOAD_TCP_TSO;
- if (priv->tunnel_en)
- info->tx_offload_capa |= (DEV_TX_OFFLOAD_OUTER_IPV4_CKSUM |
- DEV_TX_OFFLOAD_VXLAN_TNL_TSO |
- DEV_TX_OFFLOAD_GRE_TNL_TSO);
- if (priv_get_ifname(priv, &ifname) == 0)
- info->if_index = if_nametoindex(ifname);
- info->reta_size = priv->reta_idx_n ?
- priv->reta_idx_n : priv->ind_table_max_size;
- info->hash_key_size = priv->rss_conf.rss_key_len;
- info->speed_capa = priv->link_speed_capa;
- priv_unlock(priv);
-}
-
const uint32_t *
mlx5_dev_supported_ptypes_get(struct rte_eth_dev *dev)
{
};
if (dev->rx_pkt_burst == mlx5_rx_burst ||
- dev->rx_pkt_burst == mlx5_rx_burst_vec)
+ dev->rx_pkt_burst == mlx5_rx_burst_mprq ||
+ dev->rx_pkt_burst == mlx5_rx_burst_vec ||
+ dev->rx_pkt_burst == mlx5_rx_burst_mprq_vec)
return ptypes;
return NULL;
}
-/**
- * DPDK callback to retrieve physical link information.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- * @param wait_to_complete
- * Wait for request completion (ignored).
- */
-static int
-mlx5_link_update_unlocked_gset(struct rte_eth_dev *dev, int wait_to_complete)
-{
- struct priv *priv = mlx5_get_priv(dev);
- struct ethtool_cmd edata = {
- .cmd = ETHTOOL_GSET /* Deprecated since Linux v4.5. */
- };
- struct ifreq ifr;
- struct rte_eth_link dev_link;
- int link_speed = 0;
-
- /* priv_lock() is not taken to allow concurrent calls. */
-
- (void)wait_to_complete;
- if (priv_ifreq(priv, SIOCGIFFLAGS, &ifr)) {
- WARN("ioctl(SIOCGIFFLAGS) failed: %s", strerror(errno));
- return -1;
- }
- memset(&dev_link, 0, sizeof(dev_link));
- dev_link.link_status = ((ifr.ifr_flags & IFF_UP) &&
- (ifr.ifr_flags & IFF_RUNNING));
- ifr.ifr_data = (void *)&edata;
- if (priv_ifreq(priv, SIOCETHTOOL, &ifr)) {
- WARN("ioctl(SIOCETHTOOL, ETHTOOL_GSET) failed: %s",
- strerror(errno));
- return -1;
- }
- link_speed = ethtool_cmd_speed(&edata);
- if (link_speed == -1)
- dev_link.link_speed = 0;
- else
- dev_link.link_speed = link_speed;
- priv->link_speed_capa = 0;
- if (edata.supported & SUPPORTED_Autoneg)
- priv->link_speed_capa |= ETH_LINK_SPEED_AUTONEG;
- if (edata.supported & (SUPPORTED_1000baseT_Full |
- SUPPORTED_1000baseKX_Full))
- priv->link_speed_capa |= ETH_LINK_SPEED_1G;
- if (edata.supported & SUPPORTED_10000baseKR_Full)
- priv->link_speed_capa |= ETH_LINK_SPEED_10G;
- if (edata.supported & (SUPPORTED_40000baseKR4_Full |
- SUPPORTED_40000baseCR4_Full |
- SUPPORTED_40000baseSR4_Full |
- SUPPORTED_40000baseLR4_Full))
- priv->link_speed_capa |= ETH_LINK_SPEED_40G;
- dev_link.link_duplex = ((edata.duplex == DUPLEX_HALF) ?
- ETH_LINK_HALF_DUPLEX : ETH_LINK_FULL_DUPLEX);
- dev_link.link_autoneg = !(dev->data->dev_conf.link_speeds &
- ETH_LINK_SPEED_FIXED);
- if (memcmp(&dev_link, &dev->data->dev_link, sizeof(dev_link))) {
- /* Link status changed. */
- dev->data->dev_link = dev_link;
- return 0;
- }
- /* Link status is still the same. */
- return -1;
-}
-
-/**
- * Retrieve physical link information (unlocked version using new ioctl).
- *
- * @param dev
- * Pointer to Ethernet device structure.
- * @param wait_to_complete
- * Wait for request completion (ignored).
- */
-static int
-mlx5_link_update_unlocked_gs(struct rte_eth_dev *dev, int wait_to_complete)
-{
- struct priv *priv = mlx5_get_priv(dev);
- struct ethtool_link_settings gcmd = { .cmd = ETHTOOL_GLINKSETTINGS };
- struct ifreq ifr;
- struct rte_eth_link dev_link;
- uint64_t sc;
-
- (void)wait_to_complete;
- if (priv_ifreq(priv, SIOCGIFFLAGS, &ifr)) {
- WARN("ioctl(SIOCGIFFLAGS) failed: %s", strerror(errno));
- return -1;
- }
- memset(&dev_link, 0, sizeof(dev_link));
- dev_link.link_status = ((ifr.ifr_flags & IFF_UP) &&
- (ifr.ifr_flags & IFF_RUNNING));
- ifr.ifr_data = (void *)&gcmd;
- if (priv_ifreq(priv, SIOCETHTOOL, &ifr)) {
- DEBUG("ioctl(SIOCETHTOOL, ETHTOOL_GLINKSETTINGS) failed: %s",
- strerror(errno));
- return -1;
- }
- gcmd.link_mode_masks_nwords = -gcmd.link_mode_masks_nwords;
-
- alignas(struct ethtool_link_settings)
- uint8_t data[offsetof(struct ethtool_link_settings, link_mode_masks) +
- sizeof(uint32_t) * gcmd.link_mode_masks_nwords * 3];
- struct ethtool_link_settings *ecmd = (void *)data;
-
- *ecmd = gcmd;
- ifr.ifr_data = (void *)ecmd;
- if (priv_ifreq(priv, SIOCETHTOOL, &ifr)) {
- DEBUG("ioctl(SIOCETHTOOL, ETHTOOL_GLINKSETTINGS) failed: %s",
- strerror(errno));
- return -1;
- }
- dev_link.link_speed = ecmd->speed;
- sc = ecmd->link_mode_masks[0] |
- ((uint64_t)ecmd->link_mode_masks[1] << 32);
- priv->link_speed_capa = 0;
- if (sc & ETHTOOL_LINK_MODE_Autoneg_BIT)
- priv->link_speed_capa |= ETH_LINK_SPEED_AUTONEG;
- if (sc & (ETHTOOL_LINK_MODE_1000baseT_Full_BIT |
- ETHTOOL_LINK_MODE_1000baseKX_Full_BIT))
- priv->link_speed_capa |= ETH_LINK_SPEED_1G;
- if (sc & (ETHTOOL_LINK_MODE_10000baseKX4_Full_BIT |
- ETHTOOL_LINK_MODE_10000baseKR_Full_BIT |
- ETHTOOL_LINK_MODE_10000baseR_FEC_BIT))
- priv->link_speed_capa |= ETH_LINK_SPEED_10G;
- if (sc & (ETHTOOL_LINK_MODE_20000baseMLD2_Full_BIT |
- ETHTOOL_LINK_MODE_20000baseKR2_Full_BIT))
- priv->link_speed_capa |= ETH_LINK_SPEED_20G;
- if (sc & (ETHTOOL_LINK_MODE_40000baseKR4_Full_BIT |
- ETHTOOL_LINK_MODE_40000baseCR4_Full_BIT |
- ETHTOOL_LINK_MODE_40000baseSR4_Full_BIT |
- ETHTOOL_LINK_MODE_40000baseLR4_Full_BIT))
- priv->link_speed_capa |= ETH_LINK_SPEED_40G;
- if (sc & (ETHTOOL_LINK_MODE_56000baseKR4_Full_BIT |
- ETHTOOL_LINK_MODE_56000baseCR4_Full_BIT |
- ETHTOOL_LINK_MODE_56000baseSR4_Full_BIT |
- ETHTOOL_LINK_MODE_56000baseLR4_Full_BIT))
- priv->link_speed_capa |= ETH_LINK_SPEED_56G;
- if (sc & (ETHTOOL_LINK_MODE_25000baseCR_Full_BIT |
- ETHTOOL_LINK_MODE_25000baseKR_Full_BIT |
- ETHTOOL_LINK_MODE_25000baseSR_Full_BIT))
- priv->link_speed_capa |= ETH_LINK_SPEED_25G;
- if (sc & (ETHTOOL_LINK_MODE_50000baseCR2_Full_BIT |
- ETHTOOL_LINK_MODE_50000baseKR2_Full_BIT))
- priv->link_speed_capa |= ETH_LINK_SPEED_50G;
- if (sc & (ETHTOOL_LINK_MODE_100000baseKR4_Full_BIT |
- ETHTOOL_LINK_MODE_100000baseSR4_Full_BIT |
- ETHTOOL_LINK_MODE_100000baseCR4_Full_BIT |
- ETHTOOL_LINK_MODE_100000baseLR4_ER4_Full_BIT))
- priv->link_speed_capa |= ETH_LINK_SPEED_100G;
- dev_link.link_duplex = ((ecmd->duplex == DUPLEX_HALF) ?
- ETH_LINK_HALF_DUPLEX : ETH_LINK_FULL_DUPLEX);
- dev_link.link_autoneg = !(dev->data->dev_conf.link_speeds &
- ETH_LINK_SPEED_FIXED);
- if (memcmp(&dev_link, &dev->data->dev_link, sizeof(dev_link))) {
- /* Link status changed. */
- dev->data->dev_link = dev_link;
- return 0;
- }
- /* Link status is still the same. */
- return -1;
-}
-
-/**
- * DPDK callback to retrieve physical link information.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- * @param wait_to_complete
- * Wait for request completion (ignored).
- */
-int
-mlx5_link_update(struct rte_eth_dev *dev, int wait_to_complete)
-{
- struct utsname utsname;
- int ver[3];
-
- if (uname(&utsname) == -1 ||
- sscanf(utsname.release, "%d.%d.%d",
- &ver[0], &ver[1], &ver[2]) != 3 ||
- KERNEL_VERSION(ver[0], ver[1], ver[2]) < KERNEL_VERSION(4, 9, 0))
- return mlx5_link_update_unlocked_gset(dev, wait_to_complete);
- return mlx5_link_update_unlocked_gs(dev, wait_to_complete);
-}
-
/**
* DPDK callback to change the MTU.
*
* New MTU.
*
* @return
- * 0 on success, negative errno value on failure.
+ * 0 on success, a negative errno value otherwise and rte_errno is set.
*/
int
mlx5_dev_set_mtu(struct rte_eth_dev *dev, uint16_t mtu)
{
- struct priv *priv = dev->data->dev_private;
- uint16_t kern_mtu;
- int ret = 0;
-
- if (mlx5_is_secondary())
- return -E_RTE_SECONDARY;
+ struct mlx5_priv *priv = dev->data->dev_private;
+ uint16_t kern_mtu = 0;
+ int ret;
- priv_lock(priv);
- ret = priv_get_mtu(priv, &kern_mtu);
+ ret = mlx5_get_mtu(dev, &kern_mtu);
if (ret)
- goto out;
+ return ret;
/* Set kernel interface MTU first. */
- ret = priv_set_mtu(priv, mtu);
+ ret = mlx5_set_mtu(dev, mtu);
if (ret)
- goto out;
- ret = priv_get_mtu(priv, &kern_mtu);
+ return ret;
+ ret = mlx5_get_mtu(dev, &kern_mtu);
if (ret)
- goto out;
+ return ret;
if (kern_mtu == mtu) {
priv->mtu = mtu;
- DEBUG("adapter port %u MTU set to %u", priv->port, mtu);
- }
- priv_unlock(priv);
- return 0;
-out:
- ret = errno;
- WARN("cannot set port %u MTU to %u: %s", priv->port, mtu,
- strerror(ret));
- priv_unlock(priv);
- assert(ret >= 0);
- return -ret;
-}
-
-/**
- * DPDK callback to get flow control status.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- * @param[out] fc_conf
- * Flow control output buffer.
- *
- * @return
- * 0 on success, negative errno value on failure.
- */
-int
-mlx5_dev_get_flow_ctrl(struct rte_eth_dev *dev, struct rte_eth_fc_conf *fc_conf)
-{
- struct priv *priv = dev->data->dev_private;
- struct ifreq ifr;
- struct ethtool_pauseparam ethpause = {
- .cmd = ETHTOOL_GPAUSEPARAM
- };
- int ret;
-
- if (mlx5_is_secondary())
- return -E_RTE_SECONDARY;
-
- ifr.ifr_data = (void *)ðpause;
- priv_lock(priv);
- if (priv_ifreq(priv, SIOCETHTOOL, &ifr)) {
- ret = errno;
- WARN("ioctl(SIOCETHTOOL, ETHTOOL_GPAUSEPARAM)"
- " failed: %s",
- strerror(ret));
- goto out;
+ DRV_LOG(DEBUG, "port %u adapter MTU set to %u",
+ dev->data->port_id, mtu);
+ return 0;
}
-
- fc_conf->autoneg = ethpause.autoneg;
- if (ethpause.rx_pause && ethpause.tx_pause)
- fc_conf->mode = RTE_FC_FULL;
- else if (ethpause.rx_pause)
- fc_conf->mode = RTE_FC_RX_PAUSE;
- else if (ethpause.tx_pause)
- fc_conf->mode = RTE_FC_TX_PAUSE;
- else
- fc_conf->mode = RTE_FC_NONE;
- ret = 0;
-
-out:
- priv_unlock(priv);
- assert(ret >= 0);
- return -ret;
+ rte_errno = EAGAIN;
+ return -rte_errno;
}
/**
- * DPDK callback to modify flow control parameters.
+ * Configure the RX function to use.
*
* @param dev
- * Pointer to Ethernet device structure.
- * @param[in] fc_conf
- * Flow control parameters.
- *
- * @return
- * 0 on success, negative errno value on failure.
- */
-int
-mlx5_dev_set_flow_ctrl(struct rte_eth_dev *dev, struct rte_eth_fc_conf *fc_conf)
-{
- struct priv *priv = dev->data->dev_private;
- struct ifreq ifr;
- struct ethtool_pauseparam ethpause = {
- .cmd = ETHTOOL_SPAUSEPARAM
- };
- int ret;
-
- if (mlx5_is_secondary())
- return -E_RTE_SECONDARY;
-
- ifr.ifr_data = (void *)ðpause;
- ethpause.autoneg = fc_conf->autoneg;
- if (((fc_conf->mode & RTE_FC_FULL) == RTE_FC_FULL) ||
- (fc_conf->mode & RTE_FC_RX_PAUSE))
- ethpause.rx_pause = 1;
- else
- ethpause.rx_pause = 0;
-
- if (((fc_conf->mode & RTE_FC_FULL) == RTE_FC_FULL) ||
- (fc_conf->mode & RTE_FC_TX_PAUSE))
- ethpause.tx_pause = 1;
- else
- ethpause.tx_pause = 0;
-
- priv_lock(priv);
- if (priv_ifreq(priv, SIOCETHTOOL, &ifr)) {
- ret = errno;
- WARN("ioctl(SIOCETHTOOL, ETHTOOL_SPAUSEPARAM)"
- " failed: %s",
- strerror(ret));
- goto out;
- }
- ret = 0;
-
-out:
- priv_unlock(priv);
- assert(ret >= 0);
- return -ret;
-}
-
-/**
- * Get PCI information from struct ibv_device.
- *
- * @param device
- * Pointer to Ethernet device structure.
- * @param[out] pci_addr
- * PCI bus address output buffer.
+ * Pointer to private data structure.
*
* @return
- * 0 on success, -1 on failure and errno is set.
+ * Pointer to selected Rx burst function.
*/
-int
-mlx5_ibv_device_to_pci_addr(const struct ibv_device *device,
- struct rte_pci_addr *pci_addr)
+eth_rx_burst_t
+mlx5_select_rx_function(struct rte_eth_dev *dev)
{
- FILE *file;
- char line[32];
- MKSTR(path, "%s/device/uevent", device->ibdev_path);
-
- file = fopen(path, "rb");
- if (file == NULL)
- return -1;
- while (fgets(line, sizeof(line), file) == line) {
- size_t len = strlen(line);
- int ret;
+ eth_rx_burst_t rx_pkt_burst = mlx5_rx_burst;
- /* Truncate long lines. */
- if (len == (sizeof(line) - 1))
- while (line[(len - 1)] != '\n') {
- ret = fgetc(file);
- if (ret == EOF)
- break;
- line[(len - 1)] = ret;
- }
- /* Extract information. */
- if (sscanf(line,
- "PCI_SLOT_NAME="
- "%" SCNx32 ":%" SCNx8 ":%" SCNx8 ".%" SCNx8 "\n",
- &pci_addr->domain,
- &pci_addr->bus,
- &pci_addr->devid,
- &pci_addr->function) == 4) {
- ret = 0;
- break;
+ MLX5_ASSERT(dev != NULL);
+ if (mlx5_check_vec_rx_support(dev) > 0) {
+ if (mlx5_mprq_enabled(dev)) {
+ rx_pkt_burst = mlx5_rx_burst_mprq_vec;
+ DRV_LOG(DEBUG, "port %u selected vectorized"
+ " MPRQ Rx function", dev->data->port_id);
+ } else {
+ rx_pkt_burst = mlx5_rx_burst_vec;
+ DRV_LOG(DEBUG, "port %u selected vectorized"
+ " SPRQ Rx function", dev->data->port_id);
}
+ } else if (mlx5_mprq_enabled(dev)) {
+ rx_pkt_burst = mlx5_rx_burst_mprq;
+ DRV_LOG(DEBUG, "port %u selected MPRQ Rx function",
+ dev->data->port_id);
+ } else {
+ DRV_LOG(DEBUG, "port %u selected SPRQ Rx function",
+ dev->data->port_id);
}
- fclose(file);
- return 0;
+ return rx_pkt_burst;
}
/**
- * Update the link status.
+ * Get the E-Switch parameters by port id.
*
- * @param priv
- * Pointer to private structure.
+ * @param[in] port
+ * Device port id.
+ * @param[in] valid
+ * Device port id is valid, skip check. This flag is useful
+ * when trials are performed from probing and device is not
+ * flagged as valid yet (in attaching process).
+ * @param[out] es_domain_id
+ * E-Switch domain id.
+ * @param[out] es_port_id
+ * The port id of the port in the E-Switch.
*
* @return
- * Zero if the callback process can be called immediately.
+ * pointer to device private data structure containing data needed
+ * on success, NULL otherwise and rte_errno is set.
*/
-static int
-priv_link_status_update(struct priv *priv)
+struct mlx5_priv *
+mlx5_port_to_eswitch_info(uint16_t port, bool valid)
{
- struct rte_eth_link *link = &priv->dev->data->dev_link;
+ struct rte_eth_dev *dev;
+ struct mlx5_priv *priv;
- mlx5_link_update(priv->dev, 0);
- if (((link->link_speed == 0) && link->link_status) ||
- ((link->link_speed != 0) && !link->link_status)) {
- /*
- * Inconsistent status. Event likely occurred before the
- * kernel netdevice exposes the new status.
- */
- if (!priv->pending_alarm) {
- priv->pending_alarm = 1;
- rte_eal_alarm_set(MLX5_ALARM_TIMEOUT_US,
- mlx5_dev_link_status_handler,
- priv->dev);
- }
- return 1;
- } else if (unlikely(priv->pending_alarm)) {
- /* Link interrupt occurred while alarm is already scheduled. */
- priv->pending_alarm = 0;
- rte_eal_alarm_cancel(mlx5_dev_link_status_handler, priv->dev);
+ if (port >= RTE_MAX_ETHPORTS) {
+ rte_errno = EINVAL;
+ return NULL;
}
- return 0;
-}
-
-/**
- * Device status handler.
- *
- * @param priv
- * Pointer to private structure.
- * @param events
- * Pointer to event flags holder.
- *
- * @return
- * Events bitmap of callback process which can be called immediately.
- */
-static uint32_t
-priv_dev_status_handler(struct priv *priv)
-{
- struct ibv_async_event event;
- uint32_t ret = 0;
-
- /* Read all message and acknowledge them. */
- for (;;) {
- if (ibv_get_async_event(priv->ctx, &event))
- break;
- if ((event.event_type == IBV_EVENT_PORT_ACTIVE ||
- event.event_type == IBV_EVENT_PORT_ERR) &&
- (priv->dev->data->dev_conf.intr_conf.lsc == 1))
- ret |= (1 << RTE_ETH_EVENT_INTR_LSC);
- else if (event.event_type == IBV_EVENT_DEVICE_FATAL &&
- priv->dev->data->dev_conf.intr_conf.rmv == 1)
- ret |= (1 << RTE_ETH_EVENT_INTR_RMV);
- else
- DEBUG("event type %d on port %d not handled",
- event.event_type, event.element.port_num);
- ibv_ack_async_event(&event);
+ if (!valid && !rte_eth_dev_is_valid_port(port)) {
+ rte_errno = ENODEV;
+ return NULL;
}
- if (ret & (1 << RTE_ETH_EVENT_INTR_LSC))
- if (priv_link_status_update(priv))
- ret &= ~(1 << RTE_ETH_EVENT_INTR_LSC);
- return ret;
-}
-
-/**
- * Handle delayed link status event.
- *
- * @param arg
- * Registered argument.
- */
-void
-mlx5_dev_link_status_handler(void *arg)
-{
- struct rte_eth_dev *dev = arg;
- struct priv *priv = dev->data->dev_private;
- int ret;
-
- priv_lock(priv);
- assert(priv->pending_alarm == 1);
- priv->pending_alarm = 0;
- ret = priv_link_status_update(priv);
- priv_unlock(priv);
- if (!ret)
- _rte_eth_dev_callback_process(dev, RTE_ETH_EVENT_INTR_LSC, NULL,
- NULL);
-}
-
-/**
- * Handle interrupts from the NIC.
- *
- * @param[in] intr_handle
- * Interrupt handler.
- * @param cb_arg
- * Callback argument.
- */
-void
-mlx5_dev_interrupt_handler(void *cb_arg)
-{
- struct rte_eth_dev *dev = cb_arg;
- struct priv *priv = dev->data->dev_private;
- uint32_t events;
-
- priv_lock(priv);
- events = priv_dev_status_handler(priv);
- priv_unlock(priv);
- if (events & (1 << RTE_ETH_EVENT_INTR_LSC))
- _rte_eth_dev_callback_process(dev, RTE_ETH_EVENT_INTR_LSC, NULL,
- NULL);
- if (events & (1 << RTE_ETH_EVENT_INTR_RMV))
- _rte_eth_dev_callback_process(dev, RTE_ETH_EVENT_INTR_RMV, NULL,
- NULL);
-}
-
-/**
- * Handle interrupts from the socket.
- *
- * @param cb_arg
- * Callback argument.
- */
-static void
-mlx5_dev_handler_socket(void *cb_arg)
-{
- struct rte_eth_dev *dev = cb_arg;
- struct priv *priv = dev->data->dev_private;
-
- priv_lock(priv);
- priv_socket_handle(priv);
- priv_unlock(priv);
-}
-
-/**
- * Uninstall interrupt handler.
- *
- * @param priv
- * Pointer to private structure.
- * @param dev
- * Pointer to the rte_eth_dev structure.
- */
-void
-priv_dev_interrupt_handler_uninstall(struct priv *priv, struct rte_eth_dev *dev)
-{
- if (dev->data->dev_conf.intr_conf.lsc ||
- dev->data->dev_conf.intr_conf.rmv)
- rte_intr_callback_unregister(&priv->intr_handle,
- mlx5_dev_interrupt_handler, dev);
- if (priv->primary_socket)
- rte_intr_callback_unregister(&priv->intr_handle_socket,
- mlx5_dev_handler_socket, dev);
- if (priv->pending_alarm)
- rte_eal_alarm_cancel(mlx5_dev_link_status_handler, dev);
- priv->pending_alarm = 0;
- priv->intr_handle.fd = 0;
- priv->intr_handle.type = RTE_INTR_HANDLE_UNKNOWN;
- priv->intr_handle_socket.fd = 0;
- priv->intr_handle_socket.type = RTE_INTR_HANDLE_UNKNOWN;
-}
-
-/**
- * Install interrupt handler.
- *
- * @param priv
- * Pointer to private structure.
- * @param dev
- * Pointer to the rte_eth_dev structure.
- */
-void
-priv_dev_interrupt_handler_install(struct priv *priv, struct rte_eth_dev *dev)
-{
- int rc, flags;
-
- assert(!mlx5_is_secondary());
- assert(priv->ctx->async_fd > 0);
- flags = fcntl(priv->ctx->async_fd, F_GETFL);
- rc = fcntl(priv->ctx->async_fd, F_SETFL, flags | O_NONBLOCK);
- if (rc < 0) {
- INFO("failed to change file descriptor async event queue");
- dev->data->dev_conf.intr_conf.lsc = 0;
- dev->data->dev_conf.intr_conf.rmv = 0;
- }
- if (dev->data->dev_conf.intr_conf.lsc ||
- dev->data->dev_conf.intr_conf.rmv) {
- priv->intr_handle.fd = priv->ctx->async_fd;
- priv->intr_handle.type = RTE_INTR_HANDLE_EXT;
- rte_intr_callback_register(&priv->intr_handle,
- mlx5_dev_interrupt_handler, dev);
- }
-
- rc = priv_socket_init(priv);
- if (!rc && priv->primary_socket) {
- priv->intr_handle_socket.fd = priv->primary_socket;
- priv->intr_handle_socket.type = RTE_INTR_HANDLE_EXT;
- rte_intr_callback_register(&priv->intr_handle_socket,
- mlx5_dev_handler_socket, dev);
+ dev = &rte_eth_devices[port];
+ priv = dev->data->dev_private;
+ if (!(priv->representor || priv->master)) {
+ rte_errno = EINVAL;
+ return NULL;
}
+ return priv;
}
/**
- * Change the link state (UP / DOWN).
+ * Get the E-Switch parameters by device instance.
*
- * @param priv
- * Pointer to private data structure.
- * @param dev
- * Pointer to rte_eth_dev structure.
- * @param up
- * Nonzero for link up, otherwise link down.
+ * @param[in] port
+ * Device port id.
+ * @param[out] es_domain_id
+ * E-Switch domain id.
+ * @param[out] es_port_id
+ * The port id of the port in the E-Switch.
*
* @return
- * 0 on success, errno value on failure.
+ * pointer to device private data structure containing data needed
+ * on success, NULL otherwise and rte_errno is set.
*/
-static int
-priv_dev_set_link(struct priv *priv, struct rte_eth_dev *dev, int up)
+struct mlx5_priv *
+mlx5_dev_to_eswitch_info(struct rte_eth_dev *dev)
{
- int err;
+ struct mlx5_priv *priv;
- if (up) {
- err = priv_set_flags(priv, ~IFF_UP, IFF_UP);
- if (err)
- return err;
- priv_dev_select_tx_function(priv, dev);
- priv_dev_select_rx_function(priv, dev);
- } else {
- err = priv_set_flags(priv, ~IFF_UP, ~IFF_UP);
- if (err)
- return err;
- dev->rx_pkt_burst = removed_rx_burst;
- dev->tx_pkt_burst = removed_tx_burst;
+ priv = dev->data->dev_private;
+ if (!(priv->representor || priv->master)) {
+ rte_errno = EINVAL;
+ return NULL;
}
- return 0;
-}
-
-/**
- * DPDK callback to bring the link DOWN.
- *
- * @param dev
- * Pointer to Ethernet device structure.
- *
- * @return
- * 0 on success, errno value on failure.
- */
-int
-mlx5_set_link_down(struct rte_eth_dev *dev)
-{
- struct priv *priv = dev->data->dev_private;
- int err;
-
- priv_lock(priv);
- err = priv_dev_set_link(priv, dev, 0);
- priv_unlock(priv);
- return err;
+ return priv;
}
/**
- * DPDK callback to bring the link UP.
+ * DPDK callback to retrieve hairpin capabilities.
*
* @param dev
* Pointer to Ethernet device structure.
+ * @param[out] cap
+ * Storage for hairpin capability data.
*
* @return
- * 0 on success, errno value on failure.
+ * 0 on success, a negative errno value otherwise and rte_errno is set.
*/
int
-mlx5_set_link_up(struct rte_eth_dev *dev)
-{
- struct priv *priv = dev->data->dev_private;
- int err;
-
- priv_lock(priv);
- err = priv_dev_set_link(priv, dev, 1);
- priv_unlock(priv);
- return err;
-}
-
-/**
- * Configure the TX function to use.
- *
- * @param priv
- * Pointer to private data structure.
- * @param dev
- * Pointer to rte_eth_dev structure.
- */
-void
-priv_dev_select_tx_function(struct priv *priv, struct rte_eth_dev *dev)
+mlx5_hairpin_cap_get(struct rte_eth_dev *dev, struct rte_eth_hairpin_cap *cap)
{
- assert(priv != NULL);
- assert(dev != NULL);
- dev->tx_pkt_burst = mlx5_tx_burst;
- /* Select appropriate TX function. */
- if (priv->mps == MLX5_MPW_ENHANCED) {
- if (priv_check_vec_tx_support(priv) > 0) {
- if (priv_check_raw_vec_tx_support(priv) > 0)
- dev->tx_pkt_burst = mlx5_tx_burst_raw_vec;
- else
- dev->tx_pkt_burst = mlx5_tx_burst_vec;
- DEBUG("selected Enhanced MPW TX vectorized function");
- } else {
- dev->tx_pkt_burst = mlx5_tx_burst_empw;
- DEBUG("selected Enhanced MPW TX function");
- }
- } else if (priv->mps && priv->txq_inline) {
- dev->tx_pkt_burst = mlx5_tx_burst_mpw_inline;
- DEBUG("selected MPW inline TX function");
- } else if (priv->mps) {
- dev->tx_pkt_burst = mlx5_tx_burst_mpw;
- DEBUG("selected MPW TX function");
- }
-}
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct mlx5_dev_config *config = &priv->config;
-/**
- * Configure the RX function to use.
- *
- * @param priv
- * Pointer to private data structure.
- * @param dev
- * Pointer to rte_eth_dev structure.
- */
-void
-priv_dev_select_rx_function(struct priv *priv, struct rte_eth_dev *dev)
-{
- assert(priv != NULL);
- assert(dev != NULL);
- if (priv_check_vec_rx_support(priv) > 0) {
- dev->rx_pkt_burst = mlx5_rx_burst_vec;
- DEBUG("selected RX vectorized function");
- } else {
- dev->rx_pkt_burst = mlx5_rx_burst;
+ if (!priv->sh->devx || !config->dest_tir || !config->dv_flow_en) {
+ rte_errno = ENOTSUP;
+ return -rte_errno;
}
+ cap->max_nb_queues = UINT16_MAX;
+ cap->max_rx_2_tx = 1;
+ cap->max_tx_2_rx = 1;
+ cap->max_nb_desc = 8192;
+ return 0;
}