-/*-
- * 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
*/
#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 <netinet/in.h>
-#include <linux/ethtool.h>
-#include <linux/sockios.h>
-#include <fcntl.h>
-/* DPDK headers don't like -pedantic. */
-#ifdef PEDANTIC
-#pragma GCC diagnostic ignored "-pedantic"
-#endif
-#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>
-#ifdef PEDANTIC
-#pragma GCC diagnostic error "-pedantic"
-#endif
+#include <rte_string_fns.h>
+#include <rte_rwlock.h>
+#include <rte_cycles.h>
+
+#include <mlx5_malloc.h>
-#include "mlx5.h"
#include "mlx5_rxtx.h"
-#include "mlx5_utils.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)
{
- struct mlx5_secondary_data *sd;
-
- if (!mlx5_is_secondary())
- return dev->data->dev_private;
- sd = &mlx5_secondary_data[dev->data->port_id];
- return sd->data.dev_private;
+ struct mlx5_priv *priv = dev->data->dev_private;
+ unsigned int ifindex;
+
+ MLX5_ASSERT(priv);
+ MLX5_ASSERT(priv->if_index);
+ ifindex = priv->bond_ifindex > 0 ? priv->bond_ifindex : priv->if_index;
+ if (!ifindex)
+ rte_errno = ENXIO;
+ return 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_PRIMARY;
-}
-
-/**
- * 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->ctx->device->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->ctx->device->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;
-}
-/**
- * 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;
-
- MKSTR(path, "%s/device/net/%s/%s", priv->ctx->device->ibdev_path,
- ifname, entry);
-
- file = fopen(path, "rb");
- 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->ctx->device->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;
+ 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;
}
- 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;
+ 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 = 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->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;
}
- return 0;
-}
-
-/**
- * Perform ifreq ioctl() on associated Ethernet device.
- *
- * @param[in] priv
- * Pointer to private structure.
- * @param req
- * Request number to pass to ioctl().
- * @param[out] ifr
- * Interface request structure output buffer.
- *
- * @return
- * 0 on success, -1 on failure and errno is set.
- */
-int
-priv_ifreq(const struct priv *priv, int req, struct ifreq *ifr)
-{
- int sock = socket(PF_INET, SOCK_DGRAM, IPPROTO_IP);
- int ret = -1;
-
- if (sock == -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;
- if (priv_get_ifname(priv, &ifr->ifr_name) == 0)
- ret = ioctl(sock, req, ifr);
- close(sock);
- return ret;
-}
-
-/**
- * Get device MTU.
- *
- * @param priv
- * Pointer to private structure.
- * @param[out] mtu
- * MTU value output buffer.
- *
- * @return
- * 0 on success, -1 on failure and errno is set.
- */
-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;
}
/**
- * Set device MTU.
- *
- * @param priv
- * Pointer to private structure.
- * @param mtu
- * MTU value to set.
- *
- * @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.
- *
- * @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;
- return priv_set_sysfs_ulong(priv, "flags", tmp);
-}
-
-/**
- * Ethernet device configuration.
- *
- * Prepare the driver for a given number of TX and RX queues.
+ * Configure default RSS reta.
*
* @param dev
* Pointer to Ethernet device structure.
*
* @return
- * 0 on success, errno value on failure.
+ * 0 on success, a negative errno value otherwise and rte_errno is set.
*/
-static int
-dev_configure(struct rte_eth_dev *dev)
+int
+mlx5_dev_configure_rss_reta(struct rte_eth_dev *dev)
{
- struct priv *priv = dev->data->dev_private;
+ 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;
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;
- priv->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 (priv->skip_default_rss_reta)
+ return ret;
+ 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;
}
- if (rxqs_n > priv->ind_table_max_size) {
- ERROR("cannot handle this many RX queues (%u)", rxqs_n);
- return EINVAL;
+ 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;
}
- if (rxqs_n == priv->rxqs_n)
- return 0;
- INFO("%p: RX queues number update: %u -> %u",
- (void *)dev, priv->rxqs_n, rxqs_n);
+ 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((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. */
+ /*
+ * 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] = j;
- if (++j == rxqs_n)
+ (*priv->reta_idx)[i] = rss_queue_arr[j];
+ if (++j == rss_queue_n)
j = 0;
}
- return 0;
+ mlx5_free(rss_queue_arr);
+ return ret;
}
/**
- * DPDK callback for Ethernet device configuration.
+ * Sets default tuning parameters.
*
* @param dev
- * Pointer to Ethernet device structure.
- *
- * @return
- * 0 on success, negative errno value on failure.
+ * Pointer to Ethernet device.
+ * @param[out] info
+ * Info structure output buffer.
*/
-int
-mlx5_dev_configure(struct rte_eth_dev *dev)
+static void
+mlx5_set_default_params(struct rte_eth_dev *dev, struct rte_eth_dev_info *info)
{
- struct priv *priv = dev->data->dev_private;
- int ret;
-
- if (mlx5_is_secondary())
- return -E_RTE_SECONDARY;
+ 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;
+ }
+ }
+}
- priv_lock(priv);
- ret = dev_configure(dev);
- assert(ret >= 0);
- priv_unlock(priv);
- return -ret;
+/**
+ * Sets tx mbuf limiting parameters.
+ *
+ * @param dev
+ * Pointer to Ethernet device.
+ * @param[out] info
+ * Info structure output buffer.
+ */
+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;
}
/**
* @param[out] info
* Info structure output buffer.
*/
-void
+int
mlx5_dev_infos_get(struct rte_eth_dev *dev, struct rte_eth_dev_info *info)
{
- struct priv *priv = mlx5_get_priv(dev);
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct mlx5_dev_config *config = &priv->config;
unsigned int max;
- char ifname[IF_NAMESIZE];
- priv_lock(priv);
/* 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 = ((priv->device_attr.max_cq > priv->device_attr.max_qp) ?
- priv->device_attr.max_qp : priv->device_attr.max_cq);
- /* If max >= 65535 then max = 0, max_rx_queues is uint16_t. */
- if (max >= 65535)
- max = 65535;
+ 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 = 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);
- 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_get_ifname(priv, &ifname) == 0)
- info->if_index = if_nametoindex(ifname);
- /* FIXME: RETA update/query API expects the callee to know the size of
- * the indirection table, for this PMD the size varies depending on
- * the number of RX queues, it becomes impossible to find the correct
- * size if it is not fixed.
- * The API should be updated to solve this problem. */
- info->reta_size = priv->ind_table_max_size;
- info->speed_capa =
- ETH_LINK_SPEED_1G |
- ETH_LINK_SPEED_10G |
- ETH_LINK_SPEED_20G |
- ETH_LINK_SPEED_25G |
- ETH_LINK_SPEED_40G |
- ETH_LINK_SPEED_50G |
- ETH_LINK_SPEED_56G |
- ETH_LINK_SPEED_100G;
- priv_unlock(priv);
-}
-
-const uint32_t *
-mlx5_dev_supported_ptypes_get(struct rte_eth_dev *dev)
-{
- static const uint32_t ptypes[] = {
- /* refers to rxq_cq_to_pkt_type() */
- RTE_PTYPE_L3_IPV4,
- RTE_PTYPE_L3_IPV6,
- RTE_PTYPE_INNER_L3_IPV4,
- RTE_PTYPE_INNER_L3_IPV6,
- RTE_PTYPE_UNKNOWN
-
- };
-
- if (dev->rx_pkt_burst == mlx5_rx_burst ||
- dev->rx_pkt_burst == mlx5_rx_burst_sp)
- return ptypes;
- return NULL;
-}
-
-/**
- * DPDK callback to retrieve physical link information (unlocked version).
- *
- * @param dev
- * Pointer to Ethernet device structure.
- * @param wait_to_complete
- * Wait for request completion (ignored).
- */
-static int
-mlx5_link_update_unlocked(struct rte_eth_dev *dev, int wait_to_complete)
-{
- struct priv *priv = mlx5_get_priv(dev);
- struct ethtool_cmd edata = {
- .cmd = ETHTOOL_GSET
- };
- struct ifreq ifr;
- struct rte_eth_link dev_link;
- int link_speed = 0;
-
- (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;
- 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;
+ 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;
+
+ if (priv->pf_bond >= 0) {
+ /*
+ * Switch port ID is opaque value with driver defined
+ * format. Push the PF index in bonding configurations
+ * in upper four bits of port ID. If we get too many
+ * representors (more than 4K) or PFs (more than 15)
+ * this approach must be reconsidered.
+ */
+ /* Switch port ID for VF representors: 0 - 0xFFE */
+ if ((info->switch_info.port_id != 0xffff &&
+ info->switch_info.port_id >=
+ ((1 << MLX5_PORT_ID_BONDING_PF_SHIFT) - 1)) ||
+ priv->pf_bond > MLX5_PORT_ID_BONDING_PF_MASK) {
+ DRV_LOG(ERR, "can't update switch port ID"
+ " for bonding device");
+ MLX5_ASSERT(false);
+ return -ENODEV;
+ }
+ /*
+ * Switch port ID for Host PF representor
+ * (representor_id is -1) , set to 0xFFF
+ */
+ if (info->switch_info.port_id == 0xffff)
+ info->switch_info.port_id = 0xfff;
+ info->switch_info.port_id |=
+ priv->pf_bond << MLX5_PORT_ID_BONDING_PF_SHIFT;
+ }
+ 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;
+ }
}
- /* 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 priv *priv = mlx5_get_priv(dev);
- int ret;
-
- priv_lock(priv);
- ret = mlx5_link_update_unlocked(dev, wait_to_complete);
- priv_unlock(priv);
- return ret;
+ return 0;
}
/**
- * DPDK callback to change the MTU.
- *
- * Setting the MTU affects hardware MRU (packets larger than the MTU cannot be
- * received). Use this as a hint to enable/disable scattered packets support
- * and improve performance when not needed.
- * Since failure is not an option, reconfiguring queues on the fly is not
- * recommended.
+ * Get firmware version of a device.
*
* @param dev
- * Pointer to Ethernet device structure.
- * @param in_mtu
- * New MTU.
+ * 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_set_mtu(struct rte_eth_dev *dev, uint16_t mtu)
+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 = 0;
- unsigned int i;
- uint16_t (*rx_func)(void *, struct rte_mbuf **, uint16_t) =
- mlx5_rx_burst;
-
- if (mlx5_is_secondary())
- return -E_RTE_SECONDARY;
-
- priv_lock(priv);
- /* Set kernel interface MTU first. */
- if (priv_set_mtu(priv, mtu)) {
- ret = errno;
- WARN("cannot set port %u MTU to %u: %s", priv->port, mtu,
- strerror(ret));
- goto out;
- } else
- DEBUG("adapter port %u MTU set to %u", priv->port, mtu);
- priv->mtu = mtu;
- /* Temporarily replace RX handler with a fake one, assuming it has not
- * been copied elsewhere. */
- dev->rx_pkt_burst = removed_rx_burst;
- /* Make sure everyone has left mlx5_rx_burst() and uses
- * removed_rx_burst() instead. */
- rte_wmb();
- usleep(1000);
- /* Reconfigure each RX queue. */
- for (i = 0; (i != priv->rxqs_n); ++i) {
- struct rxq *rxq = (*priv->rxqs)[i];
- unsigned int max_frame_len;
- int sp;
-
- if (rxq == NULL)
- continue;
- /* Calculate new maximum frame length according to MTU and
- * toggle scattered support (sp) if necessary. */
- max_frame_len = (priv->mtu + ETHER_HDR_LEN +
- (ETHER_MAX_VLAN_FRAME_LEN - ETHER_MAX_LEN));
- sp = (max_frame_len > (rxq->mb_len - RTE_PKTMBUF_HEADROOM));
- /* Provide new values to rxq_setup(). */
- dev->data->dev_conf.rxmode.jumbo_frame = sp;
- dev->data->dev_conf.rxmode.max_rx_pkt_len = max_frame_len;
- ret = rxq_rehash(dev, rxq);
- if (ret) {
- /* Force SP RX if that queue requires it and abort. */
- if (rxq->sp)
- rx_func = mlx5_rx_burst_sp;
- break;
- }
- /* Scattered burst function takes priority. */
- if (rxq->sp)
- rx_func = mlx5_rx_burst_sp;
- }
- /* Burst functions can now be called again. */
- rte_wmb();
- dev->rx_pkt_burst = rx_func;
-out:
- priv_unlock(priv);
- assert(ret >= 0);
- return -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 (fw_size < size)
+ return size;
+ if (fw_ver != NULL)
+ strlcpy(fw_ver, attr->fw_ver, fw_size);
+ return 0;
}
/**
- * DPDK callback to get flow control status.
+ * Get supported packet types.
*
* @param dev
* Pointer to Ethernet device structure.
- * @param[out] fc_conf
- * Flow control output buffer.
*
* @return
- * 0 on success, negative errno value on failure.
+ * A pointer to the supported Packet types array.
*/
-int
-mlx5_dev_get_flow_ctrl(struct rte_eth_dev *dev, struct rte_eth_fc_conf *fc_conf)
+const uint32_t *
+mlx5_dev_supported_ptypes_get(struct rte_eth_dev *dev)
{
- struct priv *priv = dev->data->dev_private;
- struct ifreq ifr;
- struct ethtool_pauseparam ethpause = {
- .cmd = ETHTOOL_GPAUSEPARAM
+ static const uint32_t ptypes[] = {
+ /* refers to rxq_cq_to_pkt_type() */
+ RTE_PTYPE_L2_ETHER,
+ RTE_PTYPE_L3_IPV4_EXT_UNKNOWN,
+ RTE_PTYPE_L3_IPV6_EXT_UNKNOWN,
+ RTE_PTYPE_L4_NONFRAG,
+ RTE_PTYPE_L4_FRAG,
+ RTE_PTYPE_L4_TCP,
+ RTE_PTYPE_L4_UDP,
+ RTE_PTYPE_INNER_L3_IPV4_EXT_UNKNOWN,
+ RTE_PTYPE_INNER_L3_IPV6_EXT_UNKNOWN,
+ RTE_PTYPE_INNER_L4_NONFRAG,
+ RTE_PTYPE_INNER_L4_FRAG,
+ RTE_PTYPE_INNER_L4_TCP,
+ RTE_PTYPE_INNER_L4_UDP,
+ RTE_PTYPE_UNKNOWN
};
- 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;
- }
-
- 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;
+ if (dev->rx_pkt_burst == mlx5_rx_burst ||
+ 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 modify flow control parameters.
+ * DPDK callback to change the MTU.
*
* @param dev
* Pointer to Ethernet device structure.
- * @param[in] fc_conf
- * Flow control parameters.
+ * @param in_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_flow_ctrl(struct rte_eth_dev *dev, struct rte_eth_fc_conf *fc_conf)
+mlx5_dev_set_mtu(struct rte_eth_dev *dev, uint16_t mtu)
{
- struct priv *priv = dev->data->dev_private;
- struct ifreq ifr;
- struct ethtool_pauseparam ethpause = {
- .cmd = ETHTOOL_SPAUSEPARAM
- };
+ struct mlx5_priv *priv = dev->data->dev_private;
+ uint16_t kern_mtu = 0;
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 = mlx5_get_mtu(dev, &kern_mtu);
+ if (ret)
+ return ret;
+ /* Set kernel interface MTU first. */
+ ret = mlx5_set_mtu(dev, mtu);
+ if (ret)
+ return ret;
+ ret = mlx5_get_mtu(dev, &kern_mtu);
+ if (ret)
+ return ret;
+ if (kern_mtu == mtu) {
+ priv->mtu = mtu;
+ DRV_LOG(DEBUG, "port %u adapter MTU set to %u",
+ dev->data->port_id, mtu);
+ return 0;
}
- ret = 0;
-
-out:
- priv_unlock(priv);
- assert(ret >= 0);
- return -ret;
+ rte_errno = EAGAIN;
+ return -rte_errno;
}
/**
- * Get PCI information from struct ibv_device.
+ * Configure the RX function to use.
*
- * @param device
- * Pointer to Ethernet device structure.
- * @param[out] pci_addr
- * PCI bus address output buffer.
+ * @param dev
+ * 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;
-
- /* 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="
- "%" SCNx16 ":%" SCNx8 ":%" SCNx8 ".%" SCNx8 "\n",
- &pci_addr->domain,
- &pci_addr->bus,
- &pci_addr->devid,
- &pci_addr->function) == 4) {
- ret = 0;
- break;
+ eth_rx_burst_t rx_pkt_burst = mlx5_rx_burst;
+
+ 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;
}
/**
- * Link status handler.
- *
- * @param priv
- * Pointer to private structure.
- * @param dev
- * Pointer to the rte_eth_dev structure.
+ * Get the E-Switch parameters by port id.
+ *
+ * @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
- * Nonzero 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_dev_link_status_handler(struct priv *priv, struct rte_eth_dev *dev)
+struct mlx5_priv *
+mlx5_port_to_eswitch_info(uint16_t port, bool valid)
{
- struct ibv_async_event event;
- int port_change = 0;
- int ret = 0;
-
- /* Read all message and acknowledge them. */
- for (;;) {
- if (ibv_get_async_event(priv->ctx, &event))
- break;
+ struct rte_eth_dev *dev;
+ struct mlx5_priv *priv;
- if (event.event_type == IBV_EVENT_PORT_ACTIVE ||
- event.event_type == IBV_EVENT_PORT_ERR)
- port_change = 1;
- else
- DEBUG("event type %d on port %d not handled",
- event.event_type, event.element.port_num);
- ibv_ack_async_event(&event);
+ if (port >= RTE_MAX_ETHPORTS) {
+ rte_errno = EINVAL;
+ return NULL;
}
-
- if (port_change ^ priv->pending_alarm) {
- struct rte_eth_link *link = &dev->data->dev_link;
-
- priv->pending_alarm = 0;
- mlx5_link_update_unlocked(dev, 0);
- if (((link->link_speed == 0) && link->link_status) ||
- ((link->link_speed != 0) && !link->link_status)) {
- /* Inconsistent status, check again later. */
- priv->pending_alarm = 1;
- rte_eal_alarm_set(MLX5_ALARM_TIMEOUT_US,
- mlx5_dev_link_status_handler,
- dev);
- } else
- ret = 1;
+ if (!valid && !rte_eth_dev_is_valid_port(port)) {
+ rte_errno = ENODEV;
+ return NULL;
}
- 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);
- ret = priv_dev_link_status_handler(priv, dev);
- priv_unlock(priv);
- if (ret)
- _rte_eth_dev_callback_process(dev, RTE_ETH_EVENT_INTR_LSC);
-}
-
-/**
- * Handle interrupts from the NIC.
- *
- * @param[in] intr_handle
- * Interrupt handler.
- * @param cb_arg
- * Callback argument.
- */
-void
-mlx5_dev_interrupt_handler(struct rte_intr_handle *intr_handle, void *cb_arg)
-{
- struct rte_eth_dev *dev = cb_arg;
- struct priv *priv = dev->data->dev_private;
- int ret;
-
- (void)intr_handle;
- priv_lock(priv);
- ret = priv_dev_link_status_handler(priv, dev);
- priv_unlock(priv);
- if (ret)
- _rte_eth_dev_callback_process(dev, RTE_ETH_EVENT_INTR_LSC);
-}
-
-/**
- * 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)
- return;
- rte_intr_callback_unregister(&priv->intr_handle,
- mlx5_dev_interrupt_handler,
- 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 = 0;
-}
-
-/**
- * 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;
-
- if (!dev->data->dev_conf.intr_conf.lsc)
- return;
- 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;
- } else {
- 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);
+ 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 dev
- * Pointer to Ethernet device 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_set_link(struct priv *priv, int up)
+struct mlx5_priv *
+mlx5_dev_to_eswitch_info(struct rte_eth_dev *dev)
{
- struct rte_eth_dev *dev = priv->dev;
- int err;
- unsigned int i;
+ struct mlx5_priv *priv;
- if (up) {
- err = priv_set_flags(priv, ~IFF_UP, IFF_UP);
- if (err)
- return err;
- for (i = 0; i < priv->rxqs_n; i++)
- if ((*priv->rxqs)[i]->sp)
- break;
- /* Check if an sp queue exists.
- * Note: Some old frames might be received.
- */
- if (i == priv->rxqs_n)
- dev->rx_pkt_burst = mlx5_rx_burst;
- else
- dev->rx_pkt_burst = mlx5_rx_burst_sp;
- dev->tx_pkt_burst = mlx5_tx_burst;
- } 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_set_link(priv, 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)
+mlx5_hairpin_cap_get(struct rte_eth_dev *dev, struct rte_eth_hairpin_cap *cap)
{
- struct priv *priv = dev->data->dev_private;
- int err;
-
- priv_lock(priv);
- err = priv_set_link(priv, 1);
- priv_unlock(priv);
- return err;
-}
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct mlx5_dev_config *config = &priv->config;
-/**
- * Configure secondary process queues from a private data pointer (primary
- * or secondary) and update burst callbacks. Can take place only once.
- *
- * All queues must have been previously created by the primary process to
- * avoid undefined behavior.
- *
- * @param priv
- * Private data pointer from either primary or secondary process.
- *
- * @return
- * Private data pointer from secondary process, NULL in case of error.
- */
-struct priv *
-mlx5_secondary_data_setup(struct priv *priv)
-{
- unsigned int port_id = 0;
- struct mlx5_secondary_data *sd;
- void **tx_queues;
- void **rx_queues;
- unsigned int nb_tx_queues;
- unsigned int nb_rx_queues;
- unsigned int i;
-
- /* priv must be valid at this point. */
- assert(priv != NULL);
- /* priv->dev must also be valid but may point to local memory from
- * another process, possibly with the same address and must not
- * be dereferenced yet. */
- assert(priv->dev != NULL);
- /* Determine port ID by finding out where priv comes from. */
- while (1) {
- sd = &mlx5_secondary_data[port_id];
- rte_spinlock_lock(&sd->lock);
- /* Primary process? */
- if (sd->primary_priv == priv)
- break;
- /* Secondary process? */
- if (sd->data.dev_private == priv)
- break;
- rte_spinlock_unlock(&sd->lock);
- if (++port_id == RTE_DIM(mlx5_secondary_data))
- port_id = 0;
+ if (!priv->sh->devx || !config->dest_tir || !config->dv_flow_en) {
+ rte_errno = ENOTSUP;
+ return -rte_errno;
}
- /* Switch to secondary private structure. If private data has already
- * been updated by another thread, there is nothing else to do. */
- priv = sd->data.dev_private;
- if (priv->dev->data == &sd->data)
- goto end;
- /* Sanity checks. Secondary private structure is supposed to point
- * to local eth_dev, itself still pointing to the shared device data
- * structure allocated by the primary process. */
- assert(sd->shared_dev_data != &sd->data);
- assert(sd->data.nb_tx_queues == 0);
- assert(sd->data.tx_queues == NULL);
- assert(sd->data.nb_rx_queues == 0);
- assert(sd->data.rx_queues == NULL);
- assert(priv != sd->primary_priv);
- assert(priv->dev->data == sd->shared_dev_data);
- assert(priv->txqs_n == 0);
- assert(priv->txqs == NULL);
- assert(priv->rxqs_n == 0);
- assert(priv->rxqs == NULL);
- nb_tx_queues = sd->shared_dev_data->nb_tx_queues;
- nb_rx_queues = sd->shared_dev_data->nb_rx_queues;
- /* Allocate local storage for queues. */
- tx_queues = rte_zmalloc("secondary ethdev->tx_queues",
- sizeof(sd->data.tx_queues[0]) * nb_tx_queues,
- RTE_CACHE_LINE_SIZE);
- rx_queues = rte_zmalloc("secondary ethdev->rx_queues",
- sizeof(sd->data.rx_queues[0]) * nb_rx_queues,
- RTE_CACHE_LINE_SIZE);
- if (tx_queues == NULL || rx_queues == NULL)
- goto error;
- /* Lock to prevent control operations during setup. */
- priv_lock(priv);
- /* TX queues. */
- for (i = 0; i != nb_tx_queues; ++i) {
- struct txq *primary_txq = (*sd->primary_priv->txqs)[i];
- struct txq *txq;
-
- if (primary_txq == NULL)
- continue;
- txq = rte_calloc_socket("TXQ", 1, sizeof(*txq), 0,
- primary_txq->socket);
- if (txq != NULL) {
- if (txq_setup(priv->dev,
- txq,
- primary_txq->elts_n * MLX5_PMD_SGE_WR_N,
- primary_txq->socket,
- NULL) == 0) {
- txq->stats.idx = primary_txq->stats.idx;
- tx_queues[i] = txq;
- continue;
- }
- rte_free(txq);
- }
- while (i) {
- txq = tx_queues[--i];
- txq_cleanup(txq);
- rte_free(txq);
- }
- goto error;
- }
- /* RX queues. */
- for (i = 0; i != nb_rx_queues; ++i) {
- struct rxq *primary_rxq = (*sd->primary_priv->rxqs)[i];
-
- if (primary_rxq == NULL)
- continue;
- /* Not supported yet. */
- rx_queues[i] = NULL;
- }
- /* Update everything. */
- priv->txqs = (void *)tx_queues;
- priv->txqs_n = nb_tx_queues;
- priv->rxqs = (void *)rx_queues;
- priv->rxqs_n = nb_rx_queues;
- sd->data.rx_queues = rx_queues;
- sd->data.tx_queues = tx_queues;
- sd->data.nb_rx_queues = nb_rx_queues;
- sd->data.nb_tx_queues = nb_tx_queues;
- sd->data.dev_link = sd->shared_dev_data->dev_link;
- sd->data.mtu = sd->shared_dev_data->mtu;
- memcpy(sd->data.rx_queue_state, sd->shared_dev_data->rx_queue_state,
- sizeof(sd->data.rx_queue_state));
- memcpy(sd->data.tx_queue_state, sd->shared_dev_data->tx_queue_state,
- sizeof(sd->data.tx_queue_state));
- sd->data.dev_flags = sd->shared_dev_data->dev_flags;
- /* Use local data from now on. */
- rte_mb();
- priv->dev->data = &sd->data;
- rte_mb();
- priv->dev->tx_pkt_burst = mlx5_tx_burst;
- priv->dev->rx_pkt_burst = removed_rx_burst;
- priv_unlock(priv);
-end:
- /* More sanity checks. */
- assert(priv->dev->tx_pkt_burst == mlx5_tx_burst);
- assert(priv->dev->rx_pkt_burst == removed_rx_burst);
- assert(priv->dev->data == &sd->data);
- rte_spinlock_unlock(&sd->lock);
- return priv;
-error:
- priv_unlock(priv);
- rte_free(tx_queues);
- rte_free(rx_queues);
- rte_spinlock_unlock(&sd->lock);
- return NULL;
+ 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;
}