#include <rte_ether.h>
#include <rte_ethdev_driver.h>
#include <rte_flow.h>
+#include <rte_cycles.h>
#include <rte_flow_driver.h>
#include <rte_malloc.h>
#include <rte_ip.h>
#include "mlx5_defs.h"
#include "mlx5.h"
#include "mlx5_flow.h"
+#include "mlx5_flow_os.h"
#include "mlx5_rxtx.h"
-/* Dev ops structure defined in mlx5.c */
-extern const struct eth_dev_ops mlx5_dev_ops;
-extern const struct eth_dev_ops mlx5_dev_ops_isolate;
-
/** Device flow drivers. */
-#ifdef HAVE_IBV_FLOW_DV_SUPPORT
-extern const struct mlx5_flow_driver_ops mlx5_flow_dv_drv_ops;
-#endif
extern const struct mlx5_flow_driver_ops mlx5_flow_verbs_drv_ops;
const struct mlx5_flow_driver_ops mlx5_flow_null_drv_ops;
.isolate = mlx5_flow_isolate,
.query = mlx5_flow_query,
.dev_dump = mlx5_flow_dev_dump,
+ .get_aged_flows = mlx5_flow_get_aged_flows,
};
/* Convert FDIR request to Generic flow. */
*/
if (skip_mtr_reg && config->flow_mreg_c
[id + start_reg - REG_C_0] >= priv->mtr_color_reg) {
+ if (id >= (REG_C_7 - start_reg))
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM,
+ NULL, "invalid tag id");
if (config->flow_mreg_c
[id + 1 + start_reg - REG_C_0] != REG_NONE)
return config->flow_mreg_c
} flow_attr = {
.attr = {
.num_of_specs = 2,
- .port = (uint8_t)priv->ibv_port,
+ .port = (uint8_t)priv->dev_port,
},
.eth = {
.type = IBV_FLOW_SPEC_ETH,
}
}
+/**
+ * Set the Rx queue dynamic metadata (mask and offset) for a flow
+ *
+ * @param[in] dev
+ * Pointer to the Ethernet device structure.
+ */
+void
+mlx5_flow_rxq_dynf_metadata_set(struct rte_eth_dev *dev)
+{
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct mlx5_rxq_data *data;
+ unsigned int i;
+
+ for (i = 0; i != priv->rxqs_n; ++i) {
+ if (!(*priv->rxqs)[i])
+ continue;
+ data = (*priv->rxqs)[i];
+ if (!rte_flow_dynf_metadata_avail()) {
+ data->dynf_meta = 0;
+ data->flow_meta_mask = 0;
+ data->flow_meta_offset = -1;
+ } else {
+ data->dynf_meta = 1;
+ data->flow_meta_mask = rte_flow_dynf_metadata_mask;
+ data->flow_meta_offset = rte_flow_dynf_metadata_offs;
+ }
+ }
+}
+
/*
* return a pointer to the desired action in the list of actions.
*
return 0;
}
+/*
+ * Validate the default miss action.
+ *
+ * @param[in] action_flags
+ * Bit-fields that holds the actions detected until now.
+ * @param[out] error
+ * Pointer to error structure.
+ *
+ * @return
+ * 0 on success, a negative errno value otherwise and rte_errno is set.
+ */
+int
+mlx5_flow_validate_action_default_miss(uint64_t action_flags,
+ const struct rte_flow_attr *attr,
+ struct rte_flow_error *error)
+{
+ if (action_flags & MLX5_FLOW_FATE_ACTIONS)
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ACTION, NULL,
+ "can't have 2 fate actions in"
+ " same flow");
+ if (attr->egress)
+ return rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_ATTR_EGRESS, NULL,
+ "default miss action not supported "
+ "for egress");
+ if (attr->group)
+ return rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_ATTR_GROUP, NULL,
+ "only group 0 is supported");
+ if (attr->transfer)
+ return rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_ATTR_TRANSFER,
+ NULL, "transfer is not supported");
+ return 0;
+}
+
/*
* Validate the count action.
*
if (ret < 0)
return ret;
return 0;
-#endif
+#else
return rte_flow_error_set(error, ENOTSUP,
RTE_FLOW_ERROR_TYPE_ITEM, item,
"MPLS is not supported by Verbs, please"
" update.");
+#endif
}
/**
const struct rte_flow_item items[] __rte_unused,
const struct rte_flow_action actions[] __rte_unused,
bool external __rte_unused,
+ int hairpin __rte_unused,
struct rte_flow_error *error)
{
return rte_flow_error_set(error, ENOTSUP,
flow_get_drv_type(struct rte_eth_dev *dev, const struct rte_flow_attr *attr)
{
struct mlx5_priv *priv = dev->data->dev_private;
- enum mlx5_flow_drv_type type = MLX5_FLOW_TYPE_MAX;
+ /* The OS can determine first a specific flow type (DV, VERBS) */
+ enum mlx5_flow_drv_type type = mlx5_flow_os_get_type();
+ if (type != MLX5_FLOW_TYPE_MAX)
+ return type;
+ /* If no OS specific type - continue with DV/VERBS selection */
if (attr->transfer && priv->config.dv_esw_en)
type = MLX5_FLOW_TYPE_DV;
if (!attr->transfer)
* Pointer to the list of actions.
* @param[in] external
* This flow rule is created by request external to PMD.
+ * @param[in] hairpin
+ * Number of hairpin TX actions, 0 means classic flow.
* @param[out] error
* Pointer to the error structure.
*
const struct rte_flow_attr *attr,
const struct rte_flow_item items[],
const struct rte_flow_action actions[],
- bool external, struct rte_flow_error *error)
+ bool external, int hairpin, struct rte_flow_error *error)
{
const struct mlx5_flow_driver_ops *fops;
enum mlx5_flow_drv_type type = flow_get_drv_type(dev, attr);
fops = flow_get_drv_ops(type);
- return fops->validate(dev, attr, items, actions, external, error);
+ return fops->validate(dev, attr, items, actions, external,
+ hairpin, error);
}
/**
* Pointer to the list of items.
* @param[in] actions
* Pointer to the list of actions.
+ * @param[in] flow_idx
+ * This memory pool index to the flow.
* @param[out] error
* Pointer to the error structure.
*
const struct rte_flow_attr *attr,
const struct rte_flow_item items[],
const struct rte_flow_action actions[],
+ uint32_t flow_idx,
struct rte_flow_error *error)
{
const struct mlx5_flow_driver_ops *fops;
enum mlx5_flow_drv_type type = flow->drv_type;
+ struct mlx5_flow *mlx5_flow = NULL;
MLX5_ASSERT(type > MLX5_FLOW_TYPE_MIN && type < MLX5_FLOW_TYPE_MAX);
fops = flow_get_drv_ops(type);
- return fops->prepare(dev, attr, items, actions, error);
+ mlx5_flow = fops->prepare(dev, attr, items, actions, error);
+ if (mlx5_flow)
+ mlx5_flow->flow_idx = flow_idx;
+ return mlx5_flow;
}
/**
fops->destroy(dev, flow);
}
-/**
- * Validate a flow supported by the NIC.
- *
- * @see rte_flow_validate()
- * @see rte_flow_ops
- */
-int
-mlx5_flow_validate(struct rte_eth_dev *dev,
- const struct rte_flow_attr *attr,
- const struct rte_flow_item items[],
- const struct rte_flow_action actions[],
- struct rte_flow_error *error)
-{
- int ret;
-
- ret = flow_drv_validate(dev, attr, items, actions, true, error);
- if (ret < 0)
- return ret;
- return 0;
-}
-
/**
* Get RSS action from the action list.
*
/* Build a new flow. */
if (mark_id != MLX5_DEFAULT_COPY_ID) {
items[0] = (struct rte_flow_item){
- .type = MLX5_RTE_FLOW_ITEM_TYPE_TAG,
+ .type = (enum rte_flow_item_type)
+ MLX5_RTE_FLOW_ITEM_TYPE_TAG,
.spec = &tag_spec,
};
items[1] = (struct rte_flow_item){
.type = RTE_FLOW_ITEM_TYPE_END,
};
actions[0] = (struct rte_flow_action){
- .type = MLX5_RTE_FLOW_ACTION_TYPE_MARK,
+ .type = (enum rte_flow_action_type)
+ MLX5_RTE_FLOW_ACTION_TYPE_MARK,
.conf = &ftag,
};
actions[1] = (struct rte_flow_action){
- .type = MLX5_RTE_FLOW_ACTION_TYPE_COPY_MREG,
+ .type = (enum rte_flow_action_type)
+ MLX5_RTE_FLOW_ACTION_TYPE_COPY_MREG,
.conf = &cp_mreg,
};
actions[2] = (struct rte_flow_action){
.type = RTE_FLOW_ITEM_TYPE_END,
};
actions[0] = (struct rte_flow_action){
- .type = MLX5_RTE_FLOW_ACTION_TYPE_COPY_MREG,
+ .type = (enum rte_flow_action_type)
+ MLX5_RTE_FLOW_ACTION_TYPE_COPY_MREG,
.conf = &cp_mreg,
};
actions[1] = (struct rte_flow_action){
}
/* Add set meta action and end action for the Rx flow. */
tag_action = actions_rx;
- tag_action->type = MLX5_RTE_FLOW_ACTION_TYPE_TAG;
+ tag_action->type = (enum rte_flow_action_type)
+ MLX5_RTE_FLOW_ACTION_TYPE_TAG;
actions_rx++;
rte_memcpy(actions_rx, actions, sizeof(struct rte_flow_action));
actions_rx++;
rte_memcpy(actions_tx, actions, sizeof(struct rte_flow_action));
addr = (void *)&pattern_tx[2];
item = pattern_tx;
- item->type = MLX5_RTE_FLOW_ITEM_TYPE_TAG;
+ item->type = (enum rte_flow_item_type)
+ MLX5_RTE_FLOW_ITEM_TYPE_TAG;
tag_item = (void *)addr;
tag_item->data = *flow_id;
tag_item->id = mlx5_flow_get_reg_id(dev, MLX5_HAIRPIN_TX, 0, NULL);
* Associated actions (list terminated by the END action).
* @param[in] external
* This flow rule is created by request external to PMD.
+ * @param[in] flow_idx
+ * This memory pool index to the flow.
* @param[out] error
* Perform verbose error reporting if not NULL.
* @return
const struct rte_flow_attr *attr,
const struct rte_flow_item items[],
const struct rte_flow_action actions[],
- bool external, struct rte_flow_error *error)
+ bool external, uint32_t flow_idx,
+ struct rte_flow_error *error)
{
struct mlx5_flow *dev_flow;
- dev_flow = flow_drv_prepare(dev, flow, attr, items, actions, error);
+ dev_flow = flow_drv_prepare(dev, flow, attr, items, actions,
+ flow_idx, error);
if (!dev_flow)
return -rte_errno;
dev_flow->flow = flow;
case RTE_FLOW_ACTION_TYPE_METER:
/* Add the extra tag action first. */
tag_action = actions_pre;
- tag_action->type = MLX5_RTE_FLOW_ACTION_TYPE_TAG;
+ tag_action->type = (enum rte_flow_action_type)
+ MLX5_RTE_FLOW_ACTION_TYPE_TAG;
actions_pre++;
action_cur = &actions_pre;
break;
* Convert to internal match item, it is used
* for vlan push and set vid.
*/
- sfx_items->type = MLX5_RTE_FLOW_ITEM_TYPE_VLAN;
+ sfx_items->type = (enum rte_flow_item_type)
+ MLX5_RTE_FLOW_ITEM_TYPE_VLAN;
sfx_items++;
}
break;
tag_spec->id = mlx5_flow_get_reg_id(dev, MLX5_MTR_SFX, 0, &error);
tag_mask = tag_spec + 1;
tag_mask->data = 0xffffff00;
- tag_item->type = MLX5_RTE_FLOW_ITEM_TYPE_TAG;
+ tag_item->type = (enum rte_flow_item_type)
+ MLX5_RTE_FLOW_ITEM_TYPE_TAG;
tag_item->spec = tag_spec;
tag_item->last = NULL;
tag_item->mask = tag_mask;
/* Construct new actions array. */
/* Replace QUEUE/RSS action. */
split_actions[qrss_idx] = (struct rte_flow_action){
- .type = MLX5_RTE_FLOW_ACTION_TYPE_TAG,
+ .type = (enum rte_flow_action_type)
+ MLX5_RTE_FLOW_ACTION_TYPE_TAG,
.conf = set_tag,
};
}
memcpy(ext_actions, actions, sizeof(*ext_actions) * encap_idx);
if (encap_idx == actions_n - 1) {
ext_actions[actions_n - 1] = (struct rte_flow_action){
- .type = MLX5_RTE_FLOW_ACTION_TYPE_COPY_MREG,
+ .type = (enum rte_flow_action_type)
+ MLX5_RTE_FLOW_ACTION_TYPE_COPY_MREG,
.conf = cp_mreg,
};
ext_actions[actions_n] = (struct rte_flow_action){
};
} else {
ext_actions[encap_idx] = (struct rte_flow_action){
- .type = MLX5_RTE_FLOW_ACTION_TYPE_COPY_MREG,
+ .type = (enum rte_flow_action_type)
+ MLX5_RTE_FLOW_ACTION_TYPE_COPY_MREG,
.conf = cp_mreg,
};
memcpy(ext_actions + encap_idx + 1, actions + encap_idx,
* Associated actions (list terminated by the END action).
* @param[in] external
* This flow rule is created by request external to PMD.
+ * @param[in] flow_idx
+ * This memory pool index to the flow.
* @param[out] error
* Perform verbose error reporting if not NULL.
* @return
const struct rte_flow_attr *attr,
const struct rte_flow_item items[],
const struct rte_flow_action actions[],
- bool external, struct rte_flow_error *error)
+ bool external, uint32_t flow_idx,
+ struct rte_flow_error *error)
{
struct mlx5_priv *priv = dev->data->dev_private;
struct mlx5_dev_config *config = &priv->config;
!mlx5_flow_ext_mreg_supported(dev))
return flow_create_split_inner(dev, flow, NULL, prefix_layers,
attr, items, actions, external,
- error);
+ flow_idx, error);
actions_n = flow_parse_metadata_split_actions_info(actions, &qrss,
&encap_idx);
if (qrss) {
RTE_FLOW_ACTION_TYPE_VOID;
else
ext_actions[qrss - actions].type =
+ (enum rte_flow_action_type)
MLX5_RTE_FLOW_ACTION_TYPE_TAG;
/*
* Create the new actions list with removed Q/RSS action
/* Add the unmodified original or prefix subflow. */
ret = flow_create_split_inner(dev, flow, &dev_flow, prefix_layers, attr,
items, ext_actions ? ext_actions :
- actions, external, error);
+ actions, external, flow_idx, error);
if (ret < 0)
goto exit;
MLX5_ASSERT(dev_flow);
};
struct rte_flow_item q_items[] = {
{
- .type = MLX5_RTE_FLOW_ITEM_TYPE_TAG,
+ .type = (enum rte_flow_item_type)
+ MLX5_RTE_FLOW_ITEM_TYPE_TAG,
.spec = &q_tag_spec,
.last = NULL,
.mask = NULL,
ret = flow_create_split_inner(dev, flow, &dev_flow, layers,
&q_attr, mtr_sfx ? items :
q_items, q_actions,
- external, error);
+ external, flow_idx, error);
if (ret < 0)
goto exit;
/* qrss ID should be freed if failed. */
* Associated actions (list terminated by the END action).
* @param[in] external
* This flow rule is created by request external to PMD.
+ * @param[in] flow_idx
+ * This memory pool index to the flow.
* @param[out] error
* Perform verbose error reporting if not NULL.
* @return
const struct rte_flow_attr *attr,
const struct rte_flow_item items[],
const struct rte_flow_action actions[],
- bool external, struct rte_flow_error *error)
+ bool external, uint32_t flow_idx,
+ struct rte_flow_error *error)
{
struct mlx5_priv *priv = dev->data->dev_private;
struct rte_flow_action *sfx_actions = NULL;
/* Add the prefix subflow. */
ret = flow_create_split_inner(dev, flow, &dev_flow, 0, attr,
items, pre_actions, external,
- error);
+ flow_idx, error);
if (ret) {
ret = -rte_errno;
goto exit;
0, &sfx_attr,
sfx_items ? sfx_items : items,
sfx_actions ? sfx_actions : actions,
- external, error);
+ external, flow_idx, error);
exit:
if (sfx_actions)
rte_free(sfx_actions);
* Associated actions (list terminated by the END action).
* @param[in] external
* This flow rule is created by request external to PMD.
+ * @param[in] flow_idx
+ * This memory pool index to the flow.
* @param[out] error
* Perform verbose error reporting if not NULL.
* @return
const struct rte_flow_attr *attr,
const struct rte_flow_item items[],
const struct rte_flow_action actions[],
- bool external, struct rte_flow_error *error)
+ bool external, uint32_t flow_idx,
+ struct rte_flow_error *error)
{
int ret;
ret = flow_create_split_meter(dev, flow, attr, items,
- actions, external, error);
+ actions, external, flow_idx, error);
MLX5_ASSERT(ret <= 0);
return ret;
}
const struct rte_flow_action *p_actions_rx = actions;
uint32_t i;
uint32_t idx = 0;
- int hairpin_flow = 0;
+ int hairpin_flow;
uint32_t hairpin_id = 0;
struct rte_flow_attr attr_tx = { .priority = 0 };
- int ret = flow_drv_validate(dev, attr, items, p_actions_rx, external,
- error);
+ int ret;
+ hairpin_flow = flow_check_hairpin_split(dev, attr, actions);
+ ret = flow_drv_validate(dev, attr, items, p_actions_rx,
+ external, hairpin_flow, error);
if (ret < 0)
return 0;
- hairpin_flow = flow_check_hairpin_split(dev, attr, actions);
if (hairpin_flow > 0) {
if (hairpin_flow > MLX5_MAX_SPLIT_ACTIONS) {
rte_errno = EINVAL;
*/
ret = flow_create_split_outer(dev, flow, attr,
buf->entry[i].pattern,
- p_actions_rx, external,
+ p_actions_rx, external, idx,
error);
if (ret < 0)
goto error;
attr_tx.ingress = 0;
attr_tx.egress = 1;
dev_flow = flow_drv_prepare(dev, flow, &attr_tx, items_tx.items,
- actions_hairpin_tx.actions, error);
+ actions_hairpin_tx.actions,
+ idx, error);
if (!dev_flow)
goto error;
dev_flow->flow = flow;
actions, false, &error);
}
+/**
+ * Validate a flow supported by the NIC.
+ *
+ * @see rte_flow_validate()
+ * @see rte_flow_ops
+ */
+int
+mlx5_flow_validate(struct rte_eth_dev *dev,
+ const struct rte_flow_attr *attr,
+ const struct rte_flow_item items[],
+ const struct rte_flow_action actions[],
+ struct rte_flow_error *error)
+{
+ int hairpin_flow;
+
+ hairpin_flow = flow_check_hairpin_split(dev, attr, actions);
+ return flow_drv_validate(dev, attr, items, actions,
+ true, hairpin_flow, error);
+}
+
/**
* Create a flow.
*
* are not affected.
*/
if (unlikely(!dev->data->dev_started)) {
- rte_errno = ENODEV;
DRV_LOG(DEBUG, "port %u is not started when "
"inserting a flow", dev->data->port_id);
+ rte_flow_error_set(error, ENODEV,
+ RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
+ NULL,
+ "port not started");
return NULL;
}
return (void *)(uintptr_t)flow_list_create(dev, &priv->flows,
mlx5_flow_stop_default(struct rte_eth_dev *dev)
{
flow_mreg_del_default_copy_action(dev);
+ flow_rxq_flags_clear(dev);
}
/**
};
struct rte_flow_item items[] = {
{
- .type = MLX5_RTE_FLOW_ITEM_TYPE_TX_QUEUE,
+ .type = (enum rte_flow_item_type)
+ MLX5_RTE_FLOW_ITEM_TYPE_TX_QUEUE,
.spec = &queue_spec,
.last = NULL,
.mask = &queue_mask,
if (!priv->reta_idx_n || !priv->rxqs_n) {
return 0;
}
+ if (!(dev->data->dev_conf.rxmode.mq_mode & ETH_MQ_RX_RSS_FLAG))
+ action_rss.types = 0;
for (i = 0; i != priv->reta_idx_n; ++i)
queue[i] = (*priv->reta_idx)[i];
flow_idx = flow_list_create(dev, &priv->ctrl_flows,
return mlx5_ctrl_flow_vlan(dev, eth_spec, eth_mask, NULL, NULL);
}
+/**
+ * Create default miss flow rule matching lacp traffic
+ *
+ * @param dev
+ * Pointer to Ethernet device.
+ * @param eth_spec
+ * An Ethernet flow spec to apply.
+ *
+ * @return
+ * 0 on success, a negative errno value otherwise and rte_errno is set.
+ */
+int
+mlx5_flow_lacp_miss(struct rte_eth_dev *dev)
+{
+ struct mlx5_priv *priv = dev->data->dev_private;
+ /*
+ * The LACP matching is done by only using ether type since using
+ * a multicast dst mac causes kernel to give low priority to this flow.
+ */
+ static const struct rte_flow_item_eth lacp_spec = {
+ .type = RTE_BE16(0x8809),
+ };
+ static const struct rte_flow_item_eth lacp_mask = {
+ .type = 0xffff,
+ };
+ const struct rte_flow_attr attr = {
+ .ingress = 1,
+ };
+ struct rte_flow_item items[] = {
+ {
+ .type = RTE_FLOW_ITEM_TYPE_ETH,
+ .spec = &lacp_spec,
+ .mask = &lacp_mask,
+ },
+ {
+ .type = RTE_FLOW_ITEM_TYPE_END,
+ },
+ };
+ struct rte_flow_action actions[] = {
+ {
+ .type = (enum rte_flow_action_type)
+ MLX5_RTE_FLOW_ACTION_TYPE_DEFAULT_MISS,
+ },
+ {
+ .type = RTE_FLOW_ACTION_TYPE_END,
+ },
+ };
+ struct rte_flow_error error;
+ uint32_t flow_idx = flow_list_create(dev, &priv->ctrl_flows,
+ &attr, items, actions, false, &error);
+
+ if (!flow_idx)
+ return -rte_errno;
+ return 0;
+}
+
/**
* Destroy a flow.
*
}
priv->isolated = !!enable;
if (enable)
- dev->dev_ops = &mlx5_dev_ops_isolate;
+ dev->dev_ops = &mlx5_os_dev_ops_isolate;
else
- dev->dev_ops = &mlx5_dev_ops;
+ dev->dev_ops = &mlx5_os_dev_ops;
return 0;
}
#define MLX5_POOL_QUERY_FREQ_US 1000000
+/**
+ * Get number of all validate pools.
+ *
+ * @param[in] sh
+ * Pointer to mlx5_dev_ctx_shared object.
+ *
+ * @return
+ * The number of all validate pools.
+ */
+static uint32_t
+mlx5_get_all_valid_pool_count(struct mlx5_dev_ctx_shared *sh)
+{
+ int i;
+ uint32_t pools_n = 0;
+
+ for (i = 0; i < MLX5_CCONT_TYPE_MAX; ++i)
+ pools_n += rte_atomic16_read(&sh->cmng.ccont[i].n_valid);
+ return pools_n;
+}
+
/**
* Set the periodic procedure for triggering asynchronous batch queries for all
* the counter pools.
*
* @param[in] sh
- * Pointer to mlx5_ibv_shared object.
+ * Pointer to mlx5_dev_ctx_shared object.
*/
void
-mlx5_set_query_alarm(struct mlx5_ibv_shared *sh)
+mlx5_set_query_alarm(struct mlx5_dev_ctx_shared *sh)
{
- struct mlx5_pools_container *cont = MLX5_CNT_CONTAINER(sh, 0, 0);
- uint32_t pools_n = rte_atomic16_read(&cont->n_valid);
- uint32_t us;
+ uint32_t pools_n, us;
- cont = MLX5_CNT_CONTAINER(sh, 1, 0);
- pools_n += rte_atomic16_read(&cont->n_valid);
+ pools_n = mlx5_get_all_valid_pool_count(sh);
us = MLX5_POOL_QUERY_FREQ_US / pools_n;
DRV_LOG(DEBUG, "Set alarm for %u pools each %u us", pools_n, us);
if (rte_eal_alarm_set(us, mlx5_flow_query_alarm, sh)) {
void
mlx5_flow_query_alarm(void *arg)
{
- struct mlx5_ibv_shared *sh = arg;
+ struct mlx5_dev_ctx_shared *sh = arg;
struct mlx5_devx_obj *dcs;
uint16_t offset;
int ret;
uint8_t batch = sh->cmng.batch;
+ uint8_t age = sh->cmng.age;
uint16_t pool_index = sh->cmng.pool_index;
struct mlx5_pools_container *cont;
- struct mlx5_pools_container *mcont;
struct mlx5_flow_counter_pool *pool;
+ int cont_loop = MLX5_CCONT_TYPE_MAX;
if (sh->cmng.pending_queries >= MLX5_MAX_PENDING_QUERIES)
goto set_alarm;
next_container:
- cont = MLX5_CNT_CONTAINER(sh, batch, 1);
- mcont = MLX5_CNT_CONTAINER(sh, batch, 0);
- /* Check if resize was done and need to flip a container. */
- if (cont != mcont) {
- if (cont->pools) {
- /* Clean the old container. */
- rte_free(cont->pools);
- memset(cont, 0, sizeof(*cont));
- }
- rte_cio_wmb();
- /* Flip the host container. */
- sh->cmng.mhi[batch] ^= (uint8_t)2;
- cont = mcont;
- }
+ cont = MLX5_CNT_CONTAINER(sh, batch, age);
+ rte_spinlock_lock(&cont->resize_sl);
if (!cont->pools) {
- /* 2 empty containers case is unexpected. */
- if (unlikely(batch != sh->cmng.batch))
+ rte_spinlock_unlock(&cont->resize_sl);
+ /* Check if all the containers are empty. */
+ if (unlikely(--cont_loop == 0))
goto set_alarm;
batch ^= 0x1;
pool_index = 0;
+ if (batch == 0 && pool_index == 0) {
+ age ^= 0x1;
+ sh->cmng.batch = batch;
+ sh->cmng.age = age;
+ }
goto next_container;
}
pool = cont->pools[pool_index];
+ rte_spinlock_unlock(&cont->resize_sl);
if (pool->raw_hw)
/* There is a pool query in progress. */
goto set_alarm;
* should wait for a new round of query as the new arrived packets
* will not be taken into account.
*/
- rte_atomic64_add(&pool->start_query_gen, 1);
+ pool->query_gen++;
ret = mlx5_devx_cmd_flow_counter_query(dcs, 0, MLX5_COUNTERS_PER_POOL -
offset, NULL, NULL,
pool->raw_hw->mem_mng->dm->id,
sh->devx_comp,
(uint64_t)(uintptr_t)pool);
if (ret) {
- rte_atomic64_sub(&pool->start_query_gen, 1);
DRV_LOG(ERR, "Failed to trigger asynchronous query for dcs ID"
" %d", pool->min_dcs->id);
pool->raw_hw = NULL;
if (pool_index >= rte_atomic16_read(&cont->n_valid)) {
batch ^= 0x1;
pool_index = 0;
+ if (batch == 0 && pool_index == 0)
+ age ^= 0x1;
}
set_alarm:
sh->cmng.batch = batch;
sh->cmng.pool_index = pool_index;
+ sh->cmng.age = age;
mlx5_set_query_alarm(sh);
}
+/**
+ * Check and callback event for new aged flow in the counter pool
+ *
+ * @param[in] sh
+ * Pointer to mlx5_dev_ctx_shared object.
+ * @param[in] pool
+ * Pointer to Current counter pool.
+ */
+static void
+mlx5_flow_aging_check(struct mlx5_dev_ctx_shared *sh,
+ struct mlx5_flow_counter_pool *pool)
+{
+ struct mlx5_priv *priv;
+ struct mlx5_flow_counter *cnt;
+ struct mlx5_age_info *age_info;
+ struct mlx5_age_param *age_param;
+ struct mlx5_counter_stats_raw *cur = pool->raw_hw;
+ struct mlx5_counter_stats_raw *prev = pool->raw;
+ uint16_t curr = rte_rdtsc() / (rte_get_tsc_hz() / 10);
+ uint32_t i;
+
+ for (i = 0; i < MLX5_COUNTERS_PER_POOL; ++i) {
+ cnt = MLX5_POOL_GET_CNT(pool, i);
+ age_param = MLX5_CNT_TO_AGE(cnt);
+ if (rte_atomic16_read(&age_param->state) != AGE_CANDIDATE)
+ continue;
+ if (cur->data[i].hits != prev->data[i].hits) {
+ age_param->expire = curr + age_param->timeout;
+ continue;
+ }
+ if ((uint16_t)(curr - age_param->expire) >= (UINT16_MAX / 2))
+ continue;
+ /**
+ * Hold the lock first, or if between the
+ * state AGE_TMOUT and tailq operation the
+ * release happened, the release procedure
+ * may delete a non-existent tailq node.
+ */
+ priv = rte_eth_devices[age_param->port_id].data->dev_private;
+ age_info = GET_PORT_AGE_INFO(priv);
+ rte_spinlock_lock(&age_info->aged_sl);
+ /* If the cpmset fails, release happens. */
+ if (rte_atomic16_cmpset((volatile uint16_t *)
+ &age_param->state,
+ AGE_CANDIDATE,
+ AGE_TMOUT) ==
+ AGE_CANDIDATE) {
+ TAILQ_INSERT_TAIL(&age_info->aged_counters, cnt, next);
+ MLX5_AGE_SET(age_info, MLX5_AGE_EVENT_NEW);
+ }
+ rte_spinlock_unlock(&age_info->aged_sl);
+ }
+ for (i = 0; i < sh->max_port; i++) {
+ age_info = &sh->port[i].age_info;
+ if (!MLX5_AGE_GET(age_info, MLX5_AGE_EVENT_NEW))
+ continue;
+ if (MLX5_AGE_GET(age_info, MLX5_AGE_TRIGGER))
+ _rte_eth_dev_callback_process
+ (&rte_eth_devices[sh->port[i].devx_ih_port_id],
+ RTE_ETH_EVENT_FLOW_AGED, NULL);
+ age_info->flags = 0;
+ }
+}
+
/**
* Handler for the HW respond about ready values from an asynchronous batch
* query. This function is probably called by the host thread.
*
* @param[in] sh
- * The pointer to the shared IB device context.
+ * The pointer to the shared device context.
* @param[in] async_id
* The Devx async ID.
* @param[in] status
* The status of the completion.
*/
void
-mlx5_flow_async_pool_query_handle(struct mlx5_ibv_shared *sh,
+mlx5_flow_async_pool_query_handle(struct mlx5_dev_ctx_shared *sh,
uint64_t async_id, int status)
{
struct mlx5_flow_counter_pool *pool =
(struct mlx5_flow_counter_pool *)(uintptr_t)async_id;
struct mlx5_counter_stats_raw *raw_to_free;
+ uint8_t age = !!IS_AGE_POOL(pool);
+ uint8_t query_gen = pool->query_gen ^ 1;
+ struct mlx5_pools_container *cont =
+ MLX5_CNT_CONTAINER(sh, !IS_EXT_POOL(pool), age);
if (unlikely(status)) {
- rte_atomic64_sub(&pool->start_query_gen, 1);
raw_to_free = pool->raw_hw;
} else {
raw_to_free = pool->raw;
+ if (IS_AGE_POOL(pool))
+ mlx5_flow_aging_check(sh, pool);
rte_spinlock_lock(&pool->sl);
pool->raw = pool->raw_hw;
rte_spinlock_unlock(&pool->sl);
- MLX5_ASSERT(rte_atomic64_read(&pool->end_query_gen) + 1 ==
- rte_atomic64_read(&pool->start_query_gen));
- rte_atomic64_set(&pool->end_query_gen,
- rte_atomic64_read(&pool->start_query_gen));
/* Be sure the new raw counters data is updated in memory. */
rte_cio_wmb();
+ if (!TAILQ_EMPTY(&pool->counters[query_gen])) {
+ rte_spinlock_lock(&cont->csl);
+ TAILQ_CONCAT(&cont->counters,
+ &pool->counters[query_gen], next);
+ rte_spinlock_unlock(&cont->csl);
+ }
}
LIST_INSERT_HEAD(&sh->cmng.free_stat_raws, raw_to_free, next);
pool->raw_hw = NULL;
};
struct rte_flow_action actions[] = {
[0] = {
- .type = MLX5_RTE_FLOW_ACTION_TYPE_COPY_MREG,
+ .type = (enum rte_flow_action_type)
+ MLX5_RTE_FLOW_ACTION_TYPE_COPY_MREG,
.conf = &(struct mlx5_flow_action_copy_mreg){
.src = REG_C_1,
.dst = idx,
struct rte_flow_error *error __rte_unused)
{
struct mlx5_priv *priv = dev->data->dev_private;
- struct mlx5_ibv_shared *sh = priv->sh;
+ struct mlx5_dev_ctx_shared *sh = priv->sh;
return mlx5_devx_cmd_flow_dump(sh->fdb_domain, sh->rx_domain,
sh->tx_domain, file);
}
+
+/**
+ * Get aged-out flows.
+ *
+ * @param[in] dev
+ * Pointer to the Ethernet device structure.
+ * @param[in] context
+ * The address of an array of pointers to the aged-out flows contexts.
+ * @param[in] nb_countexts
+ * The length of context array pointers.
+ * @param[out] error
+ * Perform verbose error reporting if not NULL. Initialized in case of
+ * error only.
+ *
+ * @return
+ * how many contexts get in success, otherwise negative errno value.
+ * if nb_contexts is 0, return the amount of all aged contexts.
+ * if nb_contexts is not 0 , return the amount of aged flows reported
+ * in the context array.
+ */
+int
+mlx5_flow_get_aged_flows(struct rte_eth_dev *dev, void **contexts,
+ uint32_t nb_contexts, struct rte_flow_error *error)
+{
+ const struct mlx5_flow_driver_ops *fops;
+ struct rte_flow_attr attr = { .transfer = 0 };
+
+ if (flow_get_drv_type(dev, &attr) == MLX5_FLOW_TYPE_DV) {
+ fops = flow_get_drv_ops(MLX5_FLOW_TYPE_DV);
+ return fops->get_aged_flows(dev, contexts, nb_contexts,
+ error);
+ }
+ DRV_LOG(ERR,
+ "port %u get aged flows is not supported.",
+ dev->data->port_id);
+ return -ENOTSUP;
+}