#include <stddef.h>
#include <unistd.h>
#include <string.h>
-#include <assert.h>
-#include <dlfcn.h>
#include <stdint.h>
#include <stdlib.h>
#include <errno.h>
#include <rte_pci.h>
#include <rte_bus_pci.h>
#include <rte_common.h>
-#include <rte_config.h>
#include <rte_kvargs.h>
#include <rte_rwlock.h>
#include <rte_spinlock.h>
#include <rte_string_fns.h>
#include <rte_alarm.h>
+#include <mlx5_glue.h>
+#include <mlx5_devx_cmds.h>
+#include <mlx5_common.h>
+
+#include "mlx5_defs.h"
#include "mlx5.h"
#include "mlx5_utils.h"
#include "mlx5_rxtx.h"
#include "mlx5_autoconf.h"
-#include "mlx5_defs.h"
-#include "mlx5_glue.h"
#include "mlx5_mr.h"
#include "mlx5_flow.h"
+#include "rte_pmd_mlx5.h"
/* Device parameter to enable RX completion queue compression. */
#define MLX5_RXQ_CQE_COMP_EN "rxq_cqe_comp_en"
/* Device parameter to enable multi-packet send WQEs. */
#define MLX5_TXQ_MPW_EN "txq_mpw_en"
+/*
+ * Device parameter to force doorbell register mapping
+ * to non-cahed region eliminating the extra write memory barrier.
+ */
+#define MLX5_TX_DB_NC "tx_db_nc"
+
/*
* Device parameter to include 2 dsegs in the title WQEBB.
* Deprecated, ignored.
#define MLX5_FLOW_MIN_ID_POOL_SIZE 512
#define MLX5_ID_GENERATION_ARRAY_FACTOR 16
+#define MLX5_FLOW_TABLE_HLIST_ARRAY_SIZE 4096
+#define MLX5_TAGS_HLIST_ARRAY_SIZE 8192
+
/**
* Allocate ID pool structure.
*
+ * @param[in] max_id
+ * The maximum id can be allocated from the pool.
+ *
* @return
* Pointer to pool object, NULL value otherwise.
*/
struct mlx5_flow_id_pool *
-mlx5_flow_id_pool_alloc(void)
+mlx5_flow_id_pool_alloc(uint32_t max_id)
{
struct mlx5_flow_id_pool *pool;
void *mem;
pool->curr = pool->free_arr;
pool->last = pool->free_arr + MLX5_FLOW_MIN_ID_POOL_SIZE;
pool->base_index = 0;
+ pool->max_id = max_id;
return pool;
error:
rte_free(pool);
mlx5_flow_id_get(struct mlx5_flow_id_pool *pool, uint32_t *id)
{
if (pool->curr == pool->free_arr) {
- if (pool->base_index == UINT32_MAX) {
+ if (pool->base_index == pool->max_id) {
rte_errno = ENOMEM;
DRV_LOG(ERR, "no free id");
return -rte_errno;
if (pool->curr == pool->last) {
size = pool->curr - pool->free_arr;
size2 = size * MLX5_ID_GENERATION_ARRAY_FACTOR;
- assert(size2 > size);
+ MLX5_ASSERT(size2 > size);
mem = rte_malloc("", size2 * sizeof(uint32_t), 0);
if (!mem) {
DRV_LOG(ERR, "can't allocate mem for id pool");
}
#endif /* HAVE_IBV_FLOW_DV_SUPPORT */
+static int
+mlx5_config_doorbell_mapping_env(const struct mlx5_dev_config *config)
+{
+ char *env;
+ int value;
+
+ MLX5_ASSERT(rte_eal_process_type() == RTE_PROC_PRIMARY);
+ /* Get environment variable to store. */
+ env = getenv(MLX5_SHUT_UP_BF);
+ value = env ? !!strcmp(env, "0") : MLX5_ARG_UNSET;
+ if (config->dbnc == MLX5_ARG_UNSET)
+ setenv(MLX5_SHUT_UP_BF, MLX5_SHUT_UP_BF_DEFAULT, 1);
+ else
+ setenv(MLX5_SHUT_UP_BF,
+ config->dbnc == MLX5_TXDB_NCACHED ? "1" : "0", 1);
+ return value;
+}
+
+static void
+mlx5_restore_doorbell_mapping_env(int value)
+{
+ MLX5_ASSERT(rte_eal_process_type() == RTE_PROC_PRIMARY);
+ /* Restore the original environment variable state. */
+ if (value == MLX5_ARG_UNSET)
+ unsetenv(MLX5_SHUT_UP_BF);
+ else
+ setenv(MLX5_SHUT_UP_BF, value ? "1" : "0", 1);
+}
+
/**
* Allocate shared IB device context. If there is multiport device the
* master and representors will share this context, if there is single
*
* @param[in] spawn
* Pointer to the IB device attributes (name, port, etc).
+ * @param[in] config
+ * Pointer to device configuration structure.
*
* @return
* Pointer to mlx5_ibv_shared object on success,
* otherwise NULL and rte_errno is set.
*/
static struct mlx5_ibv_shared *
-mlx5_alloc_shared_ibctx(const struct mlx5_dev_spawn_data *spawn)
+mlx5_alloc_shared_ibctx(const struct mlx5_dev_spawn_data *spawn,
+ const struct mlx5_dev_config *config)
{
struct mlx5_ibv_shared *sh;
+ int dbmap_env;
int err = 0;
uint32_t i;
#ifdef HAVE_IBV_FLOW_DV_SUPPORT
struct mlx5_devx_tis_attr tis_attr = { 0 };
#endif
-assert(spawn);
+ MLX5_ASSERT(spawn);
/* Secondary process should not create the shared context. */
- assert(rte_eal_process_type() == RTE_PROC_PRIMARY);
+ MLX5_ASSERT(rte_eal_process_type() == RTE_PROC_PRIMARY);
pthread_mutex_lock(&mlx5_ibv_list_mutex);
/* Search for IB context by device name. */
LIST_FOREACH(sh, &mlx5_ibv_list, next) {
}
}
/* No device found, we have to create new shared context. */
- assert(spawn->max_port);
+ MLX5_ASSERT(spawn->max_port);
sh = rte_zmalloc("ethdev shared ib context",
sizeof(struct mlx5_ibv_shared) +
spawn->max_port *
rte_errno = ENOMEM;
goto exit;
}
+ /*
+ * Configure environment variable "MLX5_BF_SHUT_UP"
+ * before the device creation. The rdma_core library
+ * checks the variable at device creation and
+ * stores the result internally.
+ */
+ dbmap_env = mlx5_config_doorbell_mapping_env(config);
/* Try to open IB device with DV first, then usual Verbs. */
errno = 0;
sh->ctx = mlx5_glue->dv_open_device(spawn->ibv_dev);
if (sh->ctx) {
sh->devx = 1;
DRV_LOG(DEBUG, "DevX is supported");
+ /* The device is created, no need for environment. */
+ mlx5_restore_doorbell_mapping_env(dbmap_env);
} else {
+ /* The environment variable is still configured. */
sh->ctx = mlx5_glue->open_device(spawn->ibv_dev);
- if (!sh->ctx) {
- err = errno ? errno : ENODEV;
+ err = errno ? errno : ENODEV;
+ /*
+ * The environment variable is not needed anymore,
+ * all device creation attempts are completed.
+ */
+ mlx5_restore_doorbell_mapping_env(dbmap_env);
+ if (!sh->ctx)
goto error;
- }
DRV_LOG(DEBUG, "DevX is NOT supported");
}
err = mlx5_glue->query_device_ex(sh->ctx, NULL, &sh->device_attr);
goto error;
}
}
- sh->flow_id_pool = mlx5_flow_id_pool_alloc();
+ sh->flow_id_pool = mlx5_flow_id_pool_alloc(UINT32_MAX);
if (!sh->flow_id_pool) {
DRV_LOG(ERR, "can't create flow id pool");
err = ENOMEM;
return sh;
error:
pthread_mutex_unlock(&mlx5_ibv_list_mutex);
- assert(sh);
+ MLX5_ASSERT(sh);
if (sh->tis)
claim_zero(mlx5_devx_cmd_destroy(sh->tis));
if (sh->td)
if (sh->flow_id_pool)
mlx5_flow_id_pool_release(sh->flow_id_pool);
rte_free(sh);
- assert(err > 0);
+ MLX5_ASSERT(err > 0);
rte_errno = err;
return NULL;
}
mlx5_free_shared_ibctx(struct mlx5_ibv_shared *sh)
{
pthread_mutex_lock(&mlx5_ibv_list_mutex);
-#ifndef NDEBUG
+#ifdef RTE_LIBRTE_MLX5_DEBUG
/* Check the object presence in the list. */
struct mlx5_ibv_shared *lctx;
LIST_FOREACH(lctx, &mlx5_ibv_list, next)
if (lctx == sh)
break;
- assert(lctx);
+ MLX5_ASSERT(lctx);
if (lctx != sh) {
DRV_LOG(ERR, "Freeing non-existing shared IB context");
goto exit;
}
#endif
- assert(sh);
- assert(sh->refcnt);
+ MLX5_ASSERT(sh);
+ MLX5_ASSERT(sh->refcnt);
/* Secondary process should not free the shared context. */
- assert(rte_eal_process_type() == RTE_PROC_PRIMARY);
+ MLX5_ASSERT(rte_eal_process_type() == RTE_PROC_PRIMARY);
if (--sh->refcnt)
goto exit;
- /* Release created Memory Regions. */
- mlx5_mr_release(sh);
/* Remove from memory callback device list. */
rte_rwlock_write_lock(&mlx5_shared_data->mem_event_rwlock);
LIST_REMOVE(sh, mem_event_cb);
rte_rwlock_write_unlock(&mlx5_shared_data->mem_event_rwlock);
+ /* Release created Memory Regions. */
+ mlx5_mr_release(sh);
/* Remove context from the global device list. */
LIST_REMOVE(sh, next);
/*
* Only primary process handles async device events.
**/
mlx5_flow_counters_mng_close(sh);
- assert(!sh->intr_cnt);
+ MLX5_ASSERT(!sh->intr_cnt);
if (sh->intr_cnt)
mlx5_intr_callback_unregister
(&sh->intr_handle, mlx5_dev_interrupt_handler, sh);
pthread_mutex_unlock(&mlx5_ibv_list_mutex);
}
+/**
+ * Destroy table hash list and all the root entries per domain.
+ *
+ * @param[in] priv
+ * Pointer to the private device data structure.
+ */
+static void
+mlx5_free_table_hash_list(struct mlx5_priv *priv)
+{
+ struct mlx5_ibv_shared *sh = priv->sh;
+ struct mlx5_flow_tbl_data_entry *tbl_data;
+ union mlx5_flow_tbl_key table_key = {
+ {
+ .table_id = 0,
+ .reserved = 0,
+ .domain = 0,
+ .direction = 0,
+ }
+ };
+ struct mlx5_hlist_entry *pos;
+
+ if (!sh->flow_tbls)
+ return;
+ pos = mlx5_hlist_lookup(sh->flow_tbls, table_key.v64);
+ if (pos) {
+ tbl_data = container_of(pos, struct mlx5_flow_tbl_data_entry,
+ entry);
+ MLX5_ASSERT(tbl_data);
+ mlx5_hlist_remove(sh->flow_tbls, pos);
+ rte_free(tbl_data);
+ }
+ table_key.direction = 1;
+ pos = mlx5_hlist_lookup(sh->flow_tbls, table_key.v64);
+ if (pos) {
+ tbl_data = container_of(pos, struct mlx5_flow_tbl_data_entry,
+ entry);
+ MLX5_ASSERT(tbl_data);
+ mlx5_hlist_remove(sh->flow_tbls, pos);
+ rte_free(tbl_data);
+ }
+ table_key.direction = 0;
+ table_key.domain = 1;
+ pos = mlx5_hlist_lookup(sh->flow_tbls, table_key.v64);
+ if (pos) {
+ tbl_data = container_of(pos, struct mlx5_flow_tbl_data_entry,
+ entry);
+ MLX5_ASSERT(tbl_data);
+ mlx5_hlist_remove(sh->flow_tbls, pos);
+ rte_free(tbl_data);
+ }
+ mlx5_hlist_destroy(sh->flow_tbls, NULL, NULL);
+}
+
+/**
+ * Initialize flow table hash list and create the root tables entry
+ * for each domain.
+ *
+ * @param[in] priv
+ * Pointer to the private device data structure.
+ *
+ * @return
+ * Zero on success, positive error code otherwise.
+ */
+static int
+mlx5_alloc_table_hash_list(struct mlx5_priv *priv)
+{
+ struct mlx5_ibv_shared *sh = priv->sh;
+ char s[MLX5_HLIST_NAMESIZE];
+ int err = 0;
+
+ MLX5_ASSERT(sh);
+ snprintf(s, sizeof(s), "%s_flow_table", priv->sh->ibdev_name);
+ sh->flow_tbls = mlx5_hlist_create(s, MLX5_FLOW_TABLE_HLIST_ARRAY_SIZE);
+ if (!sh->flow_tbls) {
+ DRV_LOG(ERR, "flow tables with hash creation failed.\n");
+ err = ENOMEM;
+ return err;
+ }
+#ifndef HAVE_MLX5DV_DR
+ /*
+ * In case we have not DR support, the zero tables should be created
+ * because DV expect to see them even if they cannot be created by
+ * RDMA-CORE.
+ */
+ union mlx5_flow_tbl_key table_key = {
+ {
+ .table_id = 0,
+ .reserved = 0,
+ .domain = 0,
+ .direction = 0,
+ }
+ };
+ struct mlx5_flow_tbl_data_entry *tbl_data = rte_zmalloc(NULL,
+ sizeof(*tbl_data), 0);
+
+ if (!tbl_data) {
+ err = ENOMEM;
+ goto error;
+ }
+ tbl_data->entry.key = table_key.v64;
+ err = mlx5_hlist_insert(sh->flow_tbls, &tbl_data->entry);
+ if (err)
+ goto error;
+ rte_atomic32_init(&tbl_data->tbl.refcnt);
+ rte_atomic32_inc(&tbl_data->tbl.refcnt);
+ table_key.direction = 1;
+ tbl_data = rte_zmalloc(NULL, sizeof(*tbl_data), 0);
+ if (!tbl_data) {
+ err = ENOMEM;
+ goto error;
+ }
+ tbl_data->entry.key = table_key.v64;
+ err = mlx5_hlist_insert(sh->flow_tbls, &tbl_data->entry);
+ if (err)
+ goto error;
+ rte_atomic32_init(&tbl_data->tbl.refcnt);
+ rte_atomic32_inc(&tbl_data->tbl.refcnt);
+ table_key.direction = 0;
+ table_key.domain = 1;
+ tbl_data = rte_zmalloc(NULL, sizeof(*tbl_data), 0);
+ if (!tbl_data) {
+ err = ENOMEM;
+ goto error;
+ }
+ tbl_data->entry.key = table_key.v64;
+ err = mlx5_hlist_insert(sh->flow_tbls, &tbl_data->entry);
+ if (err)
+ goto error;
+ rte_atomic32_init(&tbl_data->tbl.refcnt);
+ rte_atomic32_inc(&tbl_data->tbl.refcnt);
+ return err;
+error:
+ mlx5_free_table_hash_list(priv);
+#endif /* HAVE_MLX5DV_DR */
+ return err;
+}
+
/**
* Initialize DR related data within private structure.
* Routine checks the reference counter and does actual
static int
mlx5_alloc_shared_dr(struct mlx5_priv *priv)
{
-#ifdef HAVE_MLX5DV_DR
struct mlx5_ibv_shared *sh = priv->sh;
+ char s[MLX5_HLIST_NAMESIZE];
int err = 0;
+
+ if (!sh->flow_tbls)
+ err = mlx5_alloc_table_hash_list(priv);
+ else
+ DRV_LOG(DEBUG, "sh->flow_tbls[%p] already created, reuse\n",
+ (void *)sh->flow_tbls);
+ if (err)
+ return err;
+ /* Create tags hash list table. */
+ snprintf(s, sizeof(s), "%s_tags", sh->ibdev_name);
+ sh->tag_table = mlx5_hlist_create(s, MLX5_TAGS_HLIST_ARRAY_SIZE);
+ if (!sh->tag_table) {
+ DRV_LOG(ERR, "tags with hash creation failed.\n");
+ err = ENOMEM;
+ goto error;
+ }
+#ifdef HAVE_MLX5DV_DR
void *domain;
- assert(sh);
if (sh->dv_refcnt) {
/* Shared DV/DR structures is already initialized. */
sh->dv_refcnt++;
}
#endif
sh->pop_vlan_action = mlx5_glue->dr_create_flow_action_pop_vlan();
+#endif /* HAVE_MLX5DV_DR */
sh->dv_refcnt++;
priv->dr_shared = 1;
return 0;
-
error:
- /* Rollback the created objects. */
+ /* Rollback the created objects. */
if (sh->rx_domain) {
mlx5_glue->dr_destroy_domain(sh->rx_domain);
sh->rx_domain = NULL;
mlx5_glue->destroy_flow_action(sh->pop_vlan_action);
sh->pop_vlan_action = NULL;
}
+ if (sh->tag_table) {
+ /* tags should be destroyed with flow before. */
+ mlx5_hlist_destroy(sh->tag_table, NULL, NULL);
+ sh->tag_table = NULL;
+ }
+ mlx5_free_table_hash_list(priv);
return err;
-#else
- (void)priv;
- return 0;
-#endif
}
/**
static void
mlx5_free_shared_dr(struct mlx5_priv *priv)
{
-#ifdef HAVE_MLX5DV_DR
struct mlx5_ibv_shared *sh;
if (!priv->dr_shared)
return;
priv->dr_shared = 0;
sh = priv->sh;
- assert(sh);
- assert(sh->dv_refcnt);
+ MLX5_ASSERT(sh);
+#ifdef HAVE_MLX5DV_DR
+ MLX5_ASSERT(sh->dv_refcnt);
if (sh->dv_refcnt && --sh->dv_refcnt)
return;
if (sh->rx_domain) {
sh->pop_vlan_action = NULL;
}
pthread_mutex_destroy(&sh->dv_mutex);
-#else
- (void)priv;
-#endif
+#endif /* HAVE_MLX5DV_DR */
+ if (sh->tag_table) {
+ /* tags should be destroyed with flow before. */
+ mlx5_hlist_destroy(sh->tag_table, NULL, NULL);
+ sh->tag_table = NULL;
+ }
+ mlx5_free_table_hash_list(priv);
}
/**
socket = ctrl->socket;
}
- assert(data != NULL);
+ MLX5_ASSERT(data != NULL);
ret = rte_malloc_socket(__func__, size, alignment, socket);
if (!ret && size)
rte_errno = ENOMEM;
static void
mlx5_free_verbs_buf(void *ptr, void *data __rte_unused)
{
- assert(data != NULL);
+ MLX5_ASSERT(data != NULL);
rte_free(ptr);
}
mlx5_udp_tunnel_port_add(struct rte_eth_dev *dev __rte_unused,
struct rte_eth_udp_tunnel *udp_tunnel)
{
- assert(udp_tunnel != NULL);
+ MLX5_ASSERT(udp_tunnel != NULL);
if (udp_tunnel->prot_type == RTE_TUNNEL_TYPE_VXLAN &&
udp_tunnel->udp_port == 4789)
return 0;
mlx5_dev_interrupt_handler_devx_uninstall(dev);
mlx5_traffic_disable(dev);
mlx5_flow_flush(dev, NULL);
+ mlx5_flow_meter_flush(dev, NULL);
/* Prevent crashes when queues are still in use. */
dev->rx_pkt_burst = removed_rx_burst;
dev->tx_pkt_burst = removed_tx_burst;
if (priv->reta_idx != NULL)
rte_free(priv->reta_idx);
if (priv->config.vf)
- mlx5_nl_mac_addr_flush(dev);
+ mlx5_nl_mac_addr_flush(priv->nl_socket_route, mlx5_ifindex(dev),
+ dev->data->mac_addrs,
+ MLX5_MAX_MAC_ADDRESSES, priv->mac_own);
if (priv->nl_socket_route >= 0)
close(priv->nl_socket_route);
if (priv->nl_socket_rdma >= 0)
.filter_ctrl = mlx5_dev_filter_ctrl,
.rx_descriptor_status = mlx5_rx_descriptor_status,
.tx_descriptor_status = mlx5_tx_descriptor_status,
+ .rxq_info_get = mlx5_rxq_info_get,
+ .txq_info_get = mlx5_txq_info_get,
+ .rx_burst_mode_get = mlx5_rx_burst_mode_get,
+ .tx_burst_mode_get = mlx5_tx_burst_mode_get,
.rx_queue_count = mlx5_rx_queue_count,
.rx_queue_intr_enable = mlx5_rx_intr_enable,
.rx_queue_intr_disable = mlx5_rx_intr_disable,
.dev_infos_get = mlx5_dev_infos_get,
.rx_descriptor_status = mlx5_rx_descriptor_status,
.tx_descriptor_status = mlx5_tx_descriptor_status,
+ .rxq_info_get = mlx5_rxq_info_get,
+ .txq_info_get = mlx5_txq_info_get,
+ .rx_burst_mode_get = mlx5_rx_burst_mode_get,
+ .tx_burst_mode_get = mlx5_tx_burst_mode_get,
.get_module_info = mlx5_get_module_info,
.get_module_eeprom = mlx5_get_module_eeprom,
};
.filter_ctrl = mlx5_dev_filter_ctrl,
.rx_descriptor_status = mlx5_rx_descriptor_status,
.tx_descriptor_status = mlx5_tx_descriptor_status,
+ .rxq_info_get = mlx5_rxq_info_get,
+ .txq_info_get = mlx5_txq_info_get,
+ .rx_burst_mode_get = mlx5_rx_burst_mode_get,
+ .tx_burst_mode_get = mlx5_tx_burst_mode_get,
.rx_queue_intr_enable = mlx5_rx_intr_enable,
.rx_queue_intr_disable = mlx5_rx_intr_disable,
.is_removed = mlx5_is_removed,
DRV_LOG(WARNING, "%s: deprecated parameter, ignored", key);
} else if (strcmp(MLX5_TXQ_MPW_EN, key) == 0) {
config->mps = !!tmp;
+ } else if (strcmp(MLX5_TX_DB_NC, key) == 0) {
+ if (tmp != MLX5_TXDB_CACHED &&
+ tmp != MLX5_TXDB_NCACHED &&
+ tmp != MLX5_TXDB_HEURISTIC) {
+ DRV_LOG(ERR, "invalid Tx doorbell "
+ "mapping parameter");
+ rte_errno = EINVAL;
+ return -rte_errno;
+ }
+ config->dbnc = tmp;
} else if (strcmp(MLX5_TXQ_MPW_HDR_DSEG_EN, key) == 0) {
DRV_LOG(WARNING, "%s: deprecated parameter, ignored", key);
} else if (strcmp(MLX5_TXQ_MAX_INLINE_LEN, key) == 0) {
if (tmp != MLX5_XMETA_MODE_LEGACY &&
tmp != MLX5_XMETA_MODE_META16 &&
tmp != MLX5_XMETA_MODE_META32) {
- DRV_LOG(WARNING, "invalid extensive "
- "metadata parameter");
+ DRV_LOG(ERR, "invalid extensive "
+ "metadata parameter");
rte_errno = EINVAL;
return -rte_errno;
}
config->max_dump_files_num = tmp;
} else if (strcmp(MLX5_LRO_TIMEOUT_USEC, key) == 0) {
config->lro.timeout = tmp;
+ } else if (strcmp(MLX5_CLASS_ARG_NAME, key) == 0) {
+ DRV_LOG(DEBUG, "class argument is %s.", val);
} else {
DRV_LOG(WARNING, "%s: unknown parameter", key);
rte_errno = EINVAL;
MLX5_TXQ_MPW_EN,
MLX5_TXQ_MPW_HDR_DSEG_EN,
MLX5_TXQ_MAX_INLINE_LEN,
+ MLX5_TX_DB_NC,
MLX5_TX_VEC_EN,
MLX5_RX_VEC_EN,
MLX5_L3_VXLAN_EN,
MLX5_REPRESENTOR,
MLX5_MAX_DUMP_FILES_NUM,
MLX5_LRO_TIMEOUT_USEC,
+ MLX5_CLASS_ARG_NAME,
NULL,
};
struct rte_kvargs *kvlist;
if (mlx5_init_shared_data())
return -rte_errno;
sd = mlx5_shared_data;
- assert(sd);
+ MLX5_ASSERT(sd);
rte_spinlock_lock(&sd->lock);
switch (rte_eal_process_type()) {
case RTE_PROC_PRIMARY:
default:
meta = 0;
mark = 0;
- assert(false);
+ MLX5_ASSERT(false);
break;
}
if (sh->dv_mark_mask && sh->dv_mark_mask != mark)
; /* Empty. */
/* Find the first clear bit. */
j = rte_bsf64(~page->dbr_bitmap[i]);
- assert(i < (MLX5_DBR_PER_PAGE / 64));
+ MLX5_ASSERT(i < (MLX5_DBR_PER_PAGE / 64));
page->dbr_bitmap[i] |= (1 << j);
page->dbr_count++;
*dbr_page = page;
return ret;
}
+int
+rte_pmd_mlx5_get_dyn_flag_names(char *names[], unsigned int n)
+{
+ static const char *const dynf_names[] = {
+ RTE_PMD_MLX5_FINE_GRANULARITY_INLINE,
+ RTE_MBUF_DYNFLAG_METADATA_NAME
+ };
+ unsigned int i;
+
+ if (n < RTE_DIM(dynf_names))
+ return -ENOMEM;
+ for (i = 0; i < RTE_DIM(dynf_names); i++) {
+ if (names[i] == NULL)
+ return -EINVAL;
+ strcpy(names[i], dynf_names[i]);
+ }
+ return RTE_DIM(dynf_names);
+}
+
/**
* Check sibling device configurations.
*
struct mlx5_dev_config *sh_conf = NULL;
uint16_t port_id;
- assert(sh);
+ MLX5_ASSERT(sh);
/* Nothing to compare for the single/first device. */
if (sh->refcnt == 1)
return 0;
eth_dev->tx_pkt_burst = mlx5_select_tx_function(eth_dev);
return eth_dev;
}
- sh = mlx5_alloc_shared_ibctx(spawn);
+ /*
+ * Some parameters ("tx_db_nc" in particularly) are needed in
+ * advance to create dv/verbs device context. We proceed the
+ * devargs here to get ones, and later proceed devargs again
+ * to override some hardware settings.
+ */
+ err = mlx5_args(&config, dpdk_dev->devargs);
+ if (err) {
+ err = rte_errno;
+ DRV_LOG(ERR, "failed to process device arguments: %s",
+ strerror(rte_errno));
+ goto error;
+ }
+ sh = mlx5_alloc_shared_ibctx(spawn, &config);
if (!sh)
return NULL;
config.devx = sh->devx;
tunnel_en = ((dv_attr.tunnel_offloads_caps &
MLX5DV_RAW_PACKET_CAP_TUNNELED_OFFLOAD_VXLAN) &&
(dv_attr.tunnel_offloads_caps &
- MLX5DV_RAW_PACKET_CAP_TUNNELED_OFFLOAD_GRE));
+ MLX5DV_RAW_PACKET_CAP_TUNNELED_OFFLOAD_GRE) &&
+ (dv_attr.tunnel_offloads_caps &
+ MLX5DV_RAW_PACKET_CAP_TUNNELED_OFFLOAD_GENEVE));
}
DRV_LOG(DEBUG, "tunnel offloading is %ssupported",
tunnel_en ? "" : "not ");
/* Some internal functions rely on Netlink sockets, open them now. */
priv->nl_socket_rdma = mlx5_nl_init(NETLINK_RDMA);
priv->nl_socket_route = mlx5_nl_init(NETLINK_ROUTE);
- priv->nl_sn = 0;
priv->representor = !!switch_info->representor;
priv->master = !!switch_info->master;
priv->domain_id = RTE_ETH_DEV_SWITCH_DOMAIN_ID_INVALID;
err = ENOTSUP;
goto error;
}
- } else if (devx_port.comp_mask & MLX5DV_DEVX_PORT_VPORT) {
+ }
+ if (devx_port.comp_mask & MLX5DV_DEVX_PORT_VPORT) {
priv->vport_id = devx_port.vport_num;
} else if (spawn->pf_bond >= 0) {
DRV_LOG(ERR, "can't deduce vport index for port %d"
}
own_domain_id = 1;
}
- err = mlx5_args(&config, dpdk_dev->devargs);
- if (err) {
- err = rte_errno;
- DRV_LOG(ERR, "failed to process device arguments: %s",
- strerror(rte_errno));
- goto error;
- }
+ /* Override some values set by hardware configuration. */
+ mlx5_args(&config, dpdk_dev->devargs);
err = mlx5_dev_check_sibling_config(priv, &config);
if (err)
goto error;
!defined(HAVE_IBV_DEVICE_COUNTERS_SET_V45)
DRV_LOG(DEBUG, "counters are not supported");
#endif
-#ifndef HAVE_IBV_FLOW_DV_SUPPORT
+#if !defined(HAVE_IBV_FLOW_DV_SUPPORT) || !defined(HAVE_MLX5DV_DR)
if (config.dv_flow_en) {
DRV_LOG(WARNING, "DV flow is not supported");
config.dv_flow_en = 0;
else
config.mps = config.mps ? mps : MLX5_MPW_DISABLED;
DRV_LOG(INFO, "%sMPS is %s",
- config.mps == MLX5_MPW_ENHANCED ? "enhanced " : "",
+ config.mps == MLX5_MPW_ENHANCED ? "enhanced " :
+ config.mps == MLX5_MPW ? "legacy " : "",
config.mps != MLX5_MPW_DISABLED ? "enabled" : "disabled");
if (config.cqe_comp && !cqe_comp) {
DRV_LOG(WARNING, "Rx CQE compression isn't supported");
priv->mtr_color_reg = ffs(reg_c_mask) - 1 +
REG_C_0;
priv->mtr_en = 1;
+ priv->mtr_reg_share =
+ config.hca_attr.qos.flow_meter_reg_share;
DRV_LOG(DEBUG, "The REG_C meter uses is %d",
priv->mtr_color_reg);
}
* is permanent throughout the lifetime of device. So, we may store
* the ifindex here and use the cached value further.
*/
- assert(spawn->ifindex);
+ MLX5_ASSERT(spawn->ifindex);
priv->if_index = spawn->ifindex;
eth_dev->data->dev_private = priv;
priv->dev_data = eth_dev->data;
mac.addr_bytes[0], mac.addr_bytes[1],
mac.addr_bytes[2], mac.addr_bytes[3],
mac.addr_bytes[4], mac.addr_bytes[5]);
-#ifndef NDEBUG
+#ifdef RTE_LIBRTE_MLX5_DEBUG
{
char ifname[IF_NAMESIZE];
/* Register MAC address. */
claim_zero(mlx5_mac_addr_add(eth_dev, &mac, 0, 0));
if (config.vf && config.vf_nl_en)
- mlx5_nl_mac_addr_sync(eth_dev);
+ mlx5_nl_mac_addr_sync(priv->nl_socket_route,
+ mlx5_ifindex(eth_dev),
+ eth_dev->data->mac_addrs,
+ MLX5_MAX_MAC_ADDRESSES);
TAILQ_INIT(&priv->flows);
TAILQ_INIT(&priv->ctrl_flows);
+ TAILQ_INIT(&priv->flow_meters);
+ TAILQ_INIT(&priv->flow_meter_profiles);
/* Hint libmlx5 to use PMD allocator for data plane resources */
struct mlx5dv_ctx_allocators alctr = {
.alloc = &mlx5_alloc_verbs_buf,
err = mlx5_alloc_shared_dr(priv);
if (err)
goto error;
- priv->qrss_id_pool = mlx5_flow_id_pool_alloc();
+ /*
+ * RSS id is shared with meter flow id. Meter flow id can only
+ * use the 24 MSB of the register.
+ */
+ priv->qrss_id_pool = mlx5_flow_id_pool_alloc(UINT32_MAX >>
+ MLX5_MTR_COLOR_BITS);
if (!priv->qrss_id_pool) {
DRV_LOG(ERR, "can't create flow id pool");
err = ENOMEM;
}
if (sh)
mlx5_free_shared_ibctx(sh);
- assert(err > 0);
+ MLX5_ASSERT(err > 0);
rte_errno = err;
return NULL;
}
if (!file)
return -1;
/* Use safe format to check maximal buffer length. */
- assert(atol(RTE_STR(IF_NAMESIZE)) == IF_NAMESIZE);
+ MLX5_ASSERT(atol(RTE_STR(IF_NAMESIZE)) == IF_NAMESIZE);
while (fscanf(file, "%" RTE_STR(IF_NAMESIZE) "s", ifname) == 1) {
char tmp_str[IF_NAMESIZE + 32];
struct rte_pci_addr pci_addr;
struct mlx5_dev_config dev_config;
int ret;
+ if (mlx5_class_get(pci_dev->device.devargs) != MLX5_CLASS_NET) {
+ DRV_LOG(DEBUG, "Skip probing - should be probed by other mlx5"
+ " driver.");
+ return 1;
+ }
+ if (rte_eal_process_type() == RTE_PROC_PRIMARY)
+ mlx5_pmd_socket_init();
ret = mlx5_init_once();
if (ret) {
DRV_LOG(ERR, "unable to init PMD global data: %s",
strerror(rte_errno));
return -rte_errno;
}
- assert(pci_drv == &mlx5_driver);
+ MLX5_ASSERT(pci_drv == &mlx5_driver);
errno = 0;
ibv_list = mlx5_glue->get_device_list(&ret);
if (!ibv_list) {
* it may be E-Switch master device and representors.
* We have to perform identification trough the ports.
*/
- assert(nl_rdma >= 0);
- assert(ns == 0);
- assert(nd == 1);
- assert(np);
+ MLX5_ASSERT(nl_rdma >= 0);
+ MLX5_ASSERT(ns == 0);
+ MLX5_ASSERT(nd == 1);
+ MLX5_ASSERT(np);
for (i = 1; i <= np; ++i) {
list[ns].max_port = np;
list[ns].ibv_port = i;
goto exit;
}
}
- assert(ns);
+ MLX5_ASSERT(ns);
/*
* Sort list to probe devices in natural order for users convenience
* (i.e. master first, then representors from lowest to highest ID).
dev_config = (struct mlx5_dev_config){
.hw_padding = 0,
.mps = MLX5_ARG_UNSET,
+ .dbnc = MLX5_ARG_UNSET,
.rx_vec_en = 1,
.txq_inline_max = MLX5_ARG_UNSET,
.txq_inline_min = MLX5_ARG_UNSET,
.min_rxqs_num = MLX5_MPRQ_MIN_RXQS,
},
.dv_esw_en = 1,
+ .dv_flow_en = 1,
};
/* Device specific configuration. */
switch (pci_dev->id.device_id) {
close(nl_route);
if (list)
rte_free(list);
- assert(ibv_list);
+ MLX5_ASSERT(ibv_list);
mlx5_glue->free_device_list(ibv_list);
return ret;
}
RTE_PCI_DRV_PROBE_AGAIN,
};
-#ifdef RTE_IBVERBS_LINK_DLOPEN
-
-/**
- * Suffix RTE_EAL_PMD_PATH with "-glue".
- *
- * This function performs a sanity check on RTE_EAL_PMD_PATH before
- * suffixing its last component.
- *
- * @param buf[out]
- * Output buffer, should be large enough otherwise NULL is returned.
- * @param size
- * Size of @p out.
- *
- * @return
- * Pointer to @p buf or @p NULL in case suffix cannot be appended.
- */
-static char *
-mlx5_glue_path(char *buf, size_t size)
-{
- static const char *const bad[] = { "/", ".", "..", NULL };
- const char *path = RTE_EAL_PMD_PATH;
- size_t len = strlen(path);
- size_t off;
- int i;
-
- while (len && path[len - 1] == '/')
- --len;
- for (off = len; off && path[off - 1] != '/'; --off)
- ;
- for (i = 0; bad[i]; ++i)
- if (!strncmp(path + off, bad[i], (int)(len - off)))
- goto error;
- i = snprintf(buf, size, "%.*s-glue", (int)len, path);
- if (i == -1 || (size_t)i >= size)
- goto error;
- return buf;
-error:
- DRV_LOG(ERR,
- "unable to append \"-glue\" to last component of"
- " RTE_EAL_PMD_PATH (\"" RTE_EAL_PMD_PATH "\"),"
- " please re-configure DPDK");
- return NULL;
-}
-
-/**
- * Initialization routine for run-time dependency on rdma-core.
- */
-static int
-mlx5_glue_init(void)
-{
- char glue_path[sizeof(RTE_EAL_PMD_PATH) - 1 + sizeof("-glue")];
- const char *path[] = {
- /*
- * A basic security check is necessary before trusting
- * MLX5_GLUE_PATH, which may override RTE_EAL_PMD_PATH.
- */
- (geteuid() == getuid() && getegid() == getgid() ?
- getenv("MLX5_GLUE_PATH") : NULL),
- /*
- * When RTE_EAL_PMD_PATH is set, use its glue-suffixed
- * variant, otherwise let dlopen() look up libraries on its
- * own.
- */
- (*RTE_EAL_PMD_PATH ?
- mlx5_glue_path(glue_path, sizeof(glue_path)) : ""),
- };
- unsigned int i = 0;
- void *handle = NULL;
- void **sym;
- const char *dlmsg;
-
- while (!handle && i != RTE_DIM(path)) {
- const char *end;
- size_t len;
- int ret;
-
- if (!path[i]) {
- ++i;
- continue;
- }
- end = strpbrk(path[i], ":;");
- if (!end)
- end = path[i] + strlen(path[i]);
- len = end - path[i];
- ret = 0;
- do {
- char name[ret + 1];
-
- ret = snprintf(name, sizeof(name), "%.*s%s" MLX5_GLUE,
- (int)len, path[i],
- (!len || *(end - 1) == '/') ? "" : "/");
- if (ret == -1)
- break;
- if (sizeof(name) != (size_t)ret + 1)
- continue;
- DRV_LOG(DEBUG, "looking for rdma-core glue as \"%s\"",
- name);
- handle = dlopen(name, RTLD_LAZY);
- break;
- } while (1);
- path[i] = end + 1;
- if (!*end)
- ++i;
- }
- if (!handle) {
- rte_errno = EINVAL;
- dlmsg = dlerror();
- if (dlmsg)
- DRV_LOG(WARNING, "cannot load glue library: %s", dlmsg);
- goto glue_error;
- }
- sym = dlsym(handle, "mlx5_glue");
- if (!sym || !*sym) {
- rte_errno = EINVAL;
- dlmsg = dlerror();
- if (dlmsg)
- DRV_LOG(ERR, "cannot resolve glue symbol: %s", dlmsg);
- goto glue_error;
- }
- mlx5_glue = *sym;
- return 0;
-glue_error:
- if (handle)
- dlclose(handle);
- DRV_LOG(WARNING,
- "cannot initialize PMD due to missing run-time dependency on"
- " rdma-core libraries (libibverbs, libmlx5)");
- return -rte_errno;
-}
-
-#endif
-
/**
* Driver initialization routine.
*/
mlx5_set_ptype_table();
mlx5_set_cksum_table();
mlx5_set_swp_types_table();
- /*
- * RDMAV_HUGEPAGES_SAFE tells ibv_fork_init() we intend to use
- * huge pages. Calling ibv_fork_init() during init allows
- * applications to use fork() safely for purposes other than
- * using this PMD, which is not supported in forked processes.
- */
- setenv("RDMAV_HUGEPAGES_SAFE", "1", 1);
- /* Match the size of Rx completion entry to the size of a cacheline. */
- if (RTE_CACHE_LINE_SIZE == 128)
- setenv("MLX5_CQE_SIZE", "128", 0);
- /*
- * MLX5_DEVICE_FATAL_CLEANUP tells ibv_destroy functions to
- * cleanup all the Verbs resources even when the device was removed.
- */
- setenv("MLX5_DEVICE_FATAL_CLEANUP", "1", 1);
-#ifdef RTE_IBVERBS_LINK_DLOPEN
- if (mlx5_glue_init())
- return;
- assert(mlx5_glue);
-#endif
-#ifndef NDEBUG
- /* Glue structure must not contain any NULL pointers. */
- {
- unsigned int i;
-
- for (i = 0; i != sizeof(*mlx5_glue) / sizeof(void *); ++i)
- assert(((const void *const *)mlx5_glue)[i]);
- }
-#endif
- if (strcmp(mlx5_glue->version, MLX5_GLUE_VERSION)) {
- DRV_LOG(ERR,
- "rdma-core glue \"%s\" mismatch: \"%s\" is required",
- mlx5_glue->version, MLX5_GLUE_VERSION);
- return;
- }
- mlx5_glue->fork_init();
- rte_pci_register(&mlx5_driver);
+ if (mlx5_glue)
+ rte_pci_register(&mlx5_driver);
}
RTE_PMD_EXPORT_NAME(net_mlx5, __COUNTER__);