#include <rte_common.h>
#include <rte_ether.h>
-#include <rte_eth_ctrl.h>
#include <rte_ethdev_driver.h>
#include <rte_flow.h>
#include <rte_flow_driver.h>
#include "mlx5.h"
#include "mlx5_defs.h"
-#include "mlx5_prm.h"
-#include "mlx5_glue.h"
#include "mlx5_flow.h"
+#include "mlx5_glue.h"
+#include "mlx5_prm.h"
+#include "mlx5_rxtx.h"
/* Dev ops structure defined in mlx5.c */
extern const struct eth_dev_ops mlx5_dev_ops;
#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_tcf_drv_ops;
extern const struct mlx5_flow_driver_ops mlx5_flow_verbs_drv_ops;
const struct mlx5_flow_driver_ops mlx5_flow_null_drv_ops;
#ifdef HAVE_IBV_FLOW_DV_SUPPORT
[MLX5_FLOW_TYPE_DV] = &mlx5_flow_dv_drv_ops,
#endif
- [MLX5_FLOW_TYPE_TCF] = &mlx5_flow_tcf_drv_ops,
[MLX5_FLOW_TYPE_VERBS] = &mlx5_flow_verbs_drv_ops,
[MLX5_FLOW_TYPE_MAX] = &mlx5_flow_null_drv_ops
};
.next = RTE_FLOW_EXPAND_RSS_NEXT
(MLX5_EXPANSION_OUTER_IPV4_UDP,
MLX5_EXPANSION_OUTER_IPV4_TCP,
- MLX5_EXPANSION_GRE),
+ MLX5_EXPANSION_GRE,
+ MLX5_EXPANSION_IPV4,
+ MLX5_EXPANSION_IPV6),
.type = RTE_FLOW_ITEM_TYPE_IPV4,
.rss_types = ETH_RSS_IPV4 | ETH_RSS_FRAG_IPV4 |
ETH_RSS_NONFRAG_IPV4_OTHER,
[MLX5_EXPANSION_OUTER_IPV6] = {
.next = RTE_FLOW_EXPAND_RSS_NEXT
(MLX5_EXPANSION_OUTER_IPV6_UDP,
- MLX5_EXPANSION_OUTER_IPV6_TCP),
+ MLX5_EXPANSION_OUTER_IPV6_TCP,
+ MLX5_EXPANSION_IPV4,
+ MLX5_EXPANSION_IPV6),
.type = RTE_FLOW_ITEM_TYPE_IPV6,
.rss_types = ETH_RSS_IPV6 | ETH_RSS_FRAG_IPV6 |
ETH_RSS_NONFRAG_IPV6_OTHER,
/* Convert FDIR request to Generic flow. */
struct mlx5_fdir {
struct rte_flow_attr attr;
- struct rte_flow_action actions[2];
struct rte_flow_item items[4];
struct rte_flow_item_eth l2;
struct rte_flow_item_eth l2_mask;
struct rte_flow_item_udp udp;
struct rte_flow_item_tcp tcp;
} l4_mask;
+ struct rte_flow_action actions[2];
struct rte_flow_action_queue queue;
};
/* Tunnel information. */
struct mlx5_flow_tunnel_info {
- uint32_t tunnel; /**< Tunnel bit (see MLX5_FLOW_*). */
+ uint64_t tunnel; /**< Tunnel bit (see MLX5_FLOW_*). */
uint32_t ptype; /**< Tunnel Ptype (see RTE_PTYPE_*). */
};
},
{
.tunnel = MLX5_FLOW_LAYER_MPLS | MLX5_FLOW_LAYER_OUTER_L4_UDP,
- .ptype = RTE_PTYPE_TUNNEL_MPLS_IN_GRE | RTE_PTYPE_L4_UDP,
+ .ptype = RTE_PTYPE_TUNNEL_MPLS_IN_UDP | RTE_PTYPE_L4_UDP,
},
{
.tunnel = MLX5_FLOW_LAYER_MPLS,
.ptype = RTE_PTYPE_TUNNEL_MPLS_IN_GRE,
},
+ {
+ .tunnel = MLX5_FLOW_LAYER_NVGRE,
+ .ptype = RTE_PTYPE_TUNNEL_NVGRE,
+ },
+ {
+ .tunnel = MLX5_FLOW_LAYER_IPIP,
+ .ptype = RTE_PTYPE_TUNNEL_IP,
+ },
+ {
+ .tunnel = MLX5_FLOW_LAYER_IPV6_ENCAP,
+ .ptype = RTE_PTYPE_TUNNEL_IP,
+ },
};
/**
int
mlx5_flow_discover_priorities(struct rte_eth_dev *dev)
{
+ struct mlx5_priv *priv = dev->data->dev_private;
struct {
struct ibv_flow_attr attr;
struct ibv_flow_spec_eth eth;
} flow_attr = {
.attr = {
.num_of_specs = 2,
+ .port = (uint8_t)priv->ibv_port,
},
.eth = {
.type = IBV_FLOW_SPEC_ETH,
claim_zero(mlx5_glue->destroy_flow(flow));
priority = vprio[i];
}
+ mlx5_hrxq_drop_release(dev);
switch (priority) {
case 8:
priority = RTE_DIM(priority_map_3);
rte_errno = ENOTSUP;
DRV_LOG(ERR,
"port %u verbs maximum priority: %d expected 8/16",
- dev->data->port_id, vprio[i]);
+ dev->data->port_id, priority);
return -rte_errno;
}
- mlx5_hrxq_drop_release(dev);
DRV_LOG(INFO, "port %u flow maximum priority: %d",
dev->data->port_id, priority);
return priority;
uint32_t subpriority)
{
uint32_t res = 0;
- struct priv *priv = dev->data->dev_private;
+ struct mlx5_priv *priv = dev->data->dev_private;
switch (priv->config.flow_prio) {
case RTE_DIM(priority_map_3):
* @return
* 0 on success, a negative errno value otherwise and rte_errno is set.
*/
-static int
+int
mlx5_flow_item_acceptable(const struct rte_flow_item *item,
const uint8_t *mask,
const uint8_t *nic_mask,
* Item hash fields.
*
* @return
- * The hash fileds that should be used.
+ * The hash fields that should be used.
*/
uint64_t
mlx5_flow_hashfields_adjust(struct mlx5_flow *dev_flow,
- int tunnel __rte_unused, uint32_t layer_types,
+ int tunnel __rte_unused, uint64_t layer_types,
uint64_t hash_fields)
{
struct rte_flow *flow = dev_flow->flow;
* Rx queue to update.
*/
static void
-mlx5_flow_rxq_tunnel_ptype_update(struct mlx5_rxq_ctrl *rxq_ctrl)
+flow_rxq_tunnel_ptype_update(struct mlx5_rxq_ctrl *rxq_ctrl)
{
unsigned int i;
uint32_t tunnel_ptype = 0;
}
/**
- * Set the Rx queue flags (Mark/Flag and Tunnel Ptypes) according to the flow.
+ * Set the Rx queue flags (Mark/Flag and Tunnel Ptypes) according to the devive
+ * flow.
*
* @param[in] dev
* Pointer to the Ethernet device structure.
- * @param[in] flow
- * Pointer to flow structure.
+ * @param[in] dev_flow
+ * Pointer to device flow structure.
*/
static void
-mlx5_flow_rxq_flags_set(struct rte_eth_dev *dev, struct rte_flow *flow)
+flow_drv_rxq_flags_set(struct rte_eth_dev *dev, struct mlx5_flow *dev_flow)
{
- struct priv *priv = dev->data->dev_private;
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct rte_flow *flow = dev_flow->flow;
const int mark = !!(flow->actions &
(MLX5_FLOW_ACTION_FLAG | MLX5_FLOW_ACTION_MARK));
- const int tunnel = !!(flow->layers & MLX5_FLOW_LAYER_TUNNEL);
+ const int tunnel = !!(dev_flow->layers & MLX5_FLOW_LAYER_TUNNEL);
unsigned int i;
for (i = 0; i != flow->rss.queue_num; ++i) {
/* Increase the counter matching the flow. */
for (j = 0; j != MLX5_FLOW_TUNNEL; ++j) {
- if ((tunnels_info[j].tunnel & flow->layers) ==
+ if ((tunnels_info[j].tunnel &
+ dev_flow->layers) ==
tunnels_info[j].tunnel) {
rxq_ctrl->flow_tunnels_n[j]++;
break;
}
}
- mlx5_flow_rxq_tunnel_ptype_update(rxq_ctrl);
+ flow_rxq_tunnel_ptype_update(rxq_ctrl);
}
}
}
+/**
+ * Set the Rx queue flags (Mark/Flag and Tunnel Ptypes) for a flow
+ *
+ * @param[in] dev
+ * Pointer to the Ethernet device structure.
+ * @param[in] flow
+ * Pointer to flow structure.
+ */
+static void
+flow_rxq_flags_set(struct rte_eth_dev *dev, struct rte_flow *flow)
+{
+ struct mlx5_flow *dev_flow;
+
+ LIST_FOREACH(dev_flow, &flow->dev_flows, next)
+ flow_drv_rxq_flags_set(dev, dev_flow);
+}
+
/**
* Clear the Rx queue flags (Mark/Flag and Tunnel Ptype) associated with the
- * @p flow if no other flow uses it with the same kind of request.
+ * device flow if no other flow uses it with the same kind of request.
*
* @param dev
* Pointer to Ethernet device.
- * @param[in] flow
- * Pointer to the flow.
+ * @param[in] dev_flow
+ * Pointer to the device flow.
*/
static void
-mlx5_flow_rxq_flags_trim(struct rte_eth_dev *dev, struct rte_flow *flow)
+flow_drv_rxq_flags_trim(struct rte_eth_dev *dev, struct mlx5_flow *dev_flow)
{
- struct priv *priv = dev->data->dev_private;
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct rte_flow *flow = dev_flow->flow;
const int mark = !!(flow->actions &
(MLX5_FLOW_ACTION_FLAG | MLX5_FLOW_ACTION_MARK));
- const int tunnel = !!(flow->layers & MLX5_FLOW_LAYER_TUNNEL);
+ const int tunnel = !!(dev_flow->layers & MLX5_FLOW_LAYER_TUNNEL);
unsigned int i;
assert(dev->data->dev_started);
/* Decrease the counter matching the flow. */
for (j = 0; j != MLX5_FLOW_TUNNEL; ++j) {
- if ((tunnels_info[j].tunnel & flow->layers) ==
+ if ((tunnels_info[j].tunnel &
+ dev_flow->layers) ==
tunnels_info[j].tunnel) {
rxq_ctrl->flow_tunnels_n[j]--;
break;
}
}
- mlx5_flow_rxq_tunnel_ptype_update(rxq_ctrl);
+ flow_rxq_tunnel_ptype_update(rxq_ctrl);
}
}
}
+/**
+ * Clear the Rx queue flags (Mark/Flag and Tunnel Ptype) associated with the
+ * @p flow if no other flow uses it with the same kind of request.
+ *
+ * @param dev
+ * Pointer to Ethernet device.
+ * @param[in] flow
+ * Pointer to the flow.
+ */
+static void
+flow_rxq_flags_trim(struct rte_eth_dev *dev, struct rte_flow *flow)
+{
+ struct mlx5_flow *dev_flow;
+
+ LIST_FOREACH(dev_flow, &flow->dev_flows, next)
+ flow_drv_rxq_flags_trim(dev, dev_flow);
+}
+
/**
* Clear the Mark/Flag and Tunnel ptype information in all Rx queues.
*
* Pointer to Ethernet device.
*/
static void
-mlx5_flow_rxq_flags_clear(struct rte_eth_dev *dev)
+flow_rxq_flags_clear(struct rte_eth_dev *dev)
{
- struct priv *priv = dev->data->dev_private;
+ struct mlx5_priv *priv = dev->data->dev_private;
unsigned int i;
for (i = 0; i != priv->rxqs_n; ++i) {
}
}
+/*
+ * return a pointer to the desired action in the list of actions.
+ *
+ * @param[in] actions
+ * The list of actions to search the action in.
+ * @param[in] action
+ * The action to find.
+ *
+ * @return
+ * Pointer to the action in the list, if found. NULL otherwise.
+ */
+const struct rte_flow_action *
+mlx5_flow_find_action(const struct rte_flow_action *actions,
+ enum rte_flow_action_type action)
+{
+ if (actions == NULL)
+ return NULL;
+ for (; actions->type != RTE_FLOW_ACTION_TYPE_END; actions++)
+ if (actions->type == action)
+ return actions;
+ return NULL;
+}
+
/*
* Validate the flag action.
*
* Pointer to error structure.
*
* @return
- * 0 on success, a negative errno value otherwise and rte_ernno is set.
+ * 0 on success, a negative errno value otherwise and rte_errno is set.
*/
int
mlx5_flow_validate_action_drop(uint64_t action_flags,
return rte_flow_error_set(error, EINVAL,
RTE_FLOW_ERROR_TYPE_ACTION, NULL,
"can't drop and mark in same flow");
- if (action_flags & MLX5_FLOW_FATE_ACTIONS)
+ if (action_flags & (MLX5_FLOW_FATE_ACTIONS |
+ MLX5_FLOW_FATE_ESWITCH_ACTIONS))
return rte_flow_error_set(error, EINVAL,
RTE_FLOW_ERROR_TYPE_ACTION, NULL,
"can't have 2 fate actions in"
* Pointer to error structure.
*
* @return
- * 0 on success, a negative errno value otherwise and rte_ernno is set.
+ * 0 on success, a negative errno value otherwise and rte_errno is set.
*/
int
mlx5_flow_validate_action_queue(const struct rte_flow_action *action,
const struct rte_flow_attr *attr,
struct rte_flow_error *error)
{
- struct priv *priv = dev->data->dev_private;
+ struct mlx5_priv *priv = dev->data->dev_private;
const struct rte_flow_action_queue *queue = action->conf;
if (action_flags & MLX5_FLOW_FATE_ACTIONS)
RTE_FLOW_ERROR_TYPE_ACTION, NULL,
"can't have 2 fate actions in"
" same flow");
+ if (!priv->rxqs_n)
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ACTION_CONF,
+ NULL, "No Rx queues configured");
if (queue->index >= priv->rxqs_n)
return rte_flow_error_set(error, EINVAL,
RTE_FLOW_ERROR_TYPE_ACTION_CONF,
* Pointer to the Ethernet device structure.
* @param[in] attr
* Attributes of flow that includes this action.
+ * @param[in] item_flags
+ * Items that were detected.
* @param[out] error
* Pointer to error structure.
*
* @return
- * 0 on success, a negative errno value otherwise and rte_ernno is set.
+ * 0 on success, a negative errno value otherwise and rte_errno is set.
*/
int
mlx5_flow_validate_action_rss(const struct rte_flow_action *action,
uint64_t action_flags,
struct rte_eth_dev *dev,
const struct rte_flow_attr *attr,
+ uint64_t item_flags,
struct rte_flow_error *error)
{
- struct priv *priv = dev->data->dev_private;
+ struct mlx5_priv *priv = dev->data->dev_private;
const struct rte_flow_action_rss *rss = action->conf;
+ int tunnel = !!(item_flags & MLX5_FLOW_LAYER_TUNNEL);
unsigned int i;
if (action_flags & MLX5_FLOW_FATE_ACTIONS)
RTE_FLOW_ERROR_TYPE_ACTION_CONF,
&rss->level,
"tunnel RSS is not supported");
- if (rss->key_len < MLX5_RSS_HASH_KEY_LEN)
+ /* allow RSS key_len 0 in case of NULL (default) RSS key. */
+ if (rss->key_len == 0 && rss->key != NULL)
+ return rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_ACTION_CONF,
+ &rss->key_len,
+ "RSS hash key length 0");
+ if (rss->key_len > 0 && rss->key_len < MLX5_RSS_HASH_KEY_LEN)
return rte_flow_error_set(error, ENOTSUP,
RTE_FLOW_ERROR_TYPE_ACTION_CONF,
&rss->key_len,
&rss->types,
"some RSS protocols are not"
" supported");
+ if (!priv->rxqs_n)
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ACTION_CONF,
+ NULL, "No Rx queues configured");
+ if (!rss->queue_num)
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ACTION_CONF,
+ NULL, "No queues configured");
for (i = 0; i != rss->queue_num; ++i) {
if (!(*priv->rxqs)[rss->queue[i]])
return rte_flow_error_set
RTE_FLOW_ERROR_TYPE_ATTR_EGRESS, NULL,
"rss action not supported for "
"egress");
+ if (rss->level > 1 && !tunnel)
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ACTION_CONF, NULL,
+ "inner RSS is not supported for "
+ "non-tunnel flows");
return 0;
}
* Pointer to error structure.
*
* @return
- * 0 on success, a negative errno value otherwise and rte_ernno is set.
+ * 0 on success, a negative errno value otherwise and rte_errno is set.
*/
int
-mlx5_flow_validate_action_count(struct rte_eth_dev *dev,
+mlx5_flow_validate_action_count(struct rte_eth_dev *dev __rte_unused,
const struct rte_flow_attr *attr,
struct rte_flow_error *error)
{
- struct priv *priv = dev->data->dev_private;
-
- if (!priv->config.flow_counter_en)
- return rte_flow_error_set(error, ENOTSUP,
- RTE_FLOW_ERROR_TYPE_ACTION, NULL,
- "flow counters are not supported.");
if (attr->egress)
return rte_flow_error_set(error, ENOTSUP,
RTE_FLOW_ERROR_TYPE_ATTR_EGRESS, NULL,
const struct rte_flow_attr *attributes,
struct rte_flow_error *error)
{
- struct priv *priv = dev->data->dev_private;
+ struct mlx5_priv *priv = dev->data->dev_private;
uint32_t priority_max = priv->config.flow_prio - 1;
if (attributes->group)
return rte_flow_error_set(error, ENOTSUP,
RTE_FLOW_ERROR_TYPE_ATTR_EGRESS, NULL,
"egress is not supported");
- if (attributes->transfer)
+ if (attributes->transfer && !priv->config.dv_esw_en)
return rte_flow_error_set(error, ENOTSUP,
RTE_FLOW_ERROR_TYPE_ATTR_TRANSFER,
NULL, "transfer is not supported");
return 0;
}
+/**
+ * Validate ICMP6 item.
+ *
+ * @param[in] item
+ * Item specification.
+ * @param[in] item_flags
+ * Bit-fields that holds the items 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_item_icmp6(const struct rte_flow_item *item,
+ uint64_t item_flags,
+ uint8_t target_protocol,
+ struct rte_flow_error *error)
+{
+ const struct rte_flow_item_icmp6 *mask = item->mask;
+ const int tunnel = !!(item_flags & MLX5_FLOW_LAYER_TUNNEL);
+ const uint64_t l3m = tunnel ? MLX5_FLOW_LAYER_INNER_L3_IPV6 :
+ MLX5_FLOW_LAYER_OUTER_L3_IPV6;
+ const uint64_t l4m = tunnel ? MLX5_FLOW_LAYER_INNER_L4 :
+ MLX5_FLOW_LAYER_OUTER_L4;
+ int ret;
+
+ if (target_protocol != 0xFF && target_protocol != IPPROTO_ICMPV6)
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "protocol filtering not compatible"
+ " with ICMP6 layer");
+ if (!(item_flags & l3m))
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "IPv6 is mandatory to filter on"
+ " ICMP6");
+ if (item_flags & l4m)
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "multiple L4 layers not supported");
+ if (!mask)
+ mask = &rte_flow_item_icmp6_mask;
+ ret = mlx5_flow_item_acceptable
+ (item, (const uint8_t *)mask,
+ (const uint8_t *)&rte_flow_item_icmp6_mask,
+ sizeof(struct rte_flow_item_icmp6), error);
+ if (ret < 0)
+ return ret;
+ return 0;
+}
+
+/**
+ * Validate ICMP item.
+ *
+ * @param[in] item
+ * Item specification.
+ * @param[in] item_flags
+ * Bit-fields that holds the items 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_item_icmp(const struct rte_flow_item *item,
+ uint64_t item_flags,
+ uint8_t target_protocol,
+ struct rte_flow_error *error)
+{
+ const struct rte_flow_item_icmp *mask = item->mask;
+ const int tunnel = !!(item_flags & MLX5_FLOW_LAYER_TUNNEL);
+ const uint64_t l3m = tunnel ? MLX5_FLOW_LAYER_INNER_L3_IPV4 :
+ MLX5_FLOW_LAYER_OUTER_L3_IPV4;
+ const uint64_t l4m = tunnel ? MLX5_FLOW_LAYER_INNER_L4 :
+ MLX5_FLOW_LAYER_OUTER_L4;
+ int ret;
+
+ if (target_protocol != 0xFF && target_protocol != IPPROTO_ICMP)
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "protocol filtering not compatible"
+ " with ICMP layer");
+ if (!(item_flags & l3m))
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "IPv4 is mandatory to filter"
+ " on ICMP");
+ if (item_flags & l4m)
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "multiple L4 layers not supported");
+ if (!mask)
+ mask = &rte_flow_item_icmp_mask;
+ ret = mlx5_flow_item_acceptable
+ (item, (const uint8_t *)mask,
+ (const uint8_t *)&rte_flow_item_icmp_mask,
+ sizeof(struct rte_flow_item_icmp), error);
+ if (ret < 0)
+ return ret;
+ return 0;
+}
+
/**
* Validate Ethernet item.
*
};
int ret;
int tunnel = !!(item_flags & MLX5_FLOW_LAYER_TUNNEL);
+ const uint64_t ethm = tunnel ? MLX5_FLOW_LAYER_INNER_L2 :
+ MLX5_FLOW_LAYER_OUTER_L2;
- if (item_flags & MLX5_FLOW_LAYER_OUTER_L2)
+ if (item_flags & ethm)
return rte_flow_error_set(error, ENOTSUP,
RTE_FLOW_ERROR_TYPE_ITEM, item,
- "3 levels of l2 are not supported");
- if ((item_flags & MLX5_FLOW_LAYER_INNER_L2) && !tunnel)
- return rte_flow_error_set(error, ENOTSUP,
- RTE_FLOW_ERROR_TYPE_ITEM, item,
- "2 L2 without tunnel are not supported");
+ "multiple L2 layers not supported");
if (!mask)
mask = &rte_flow_item_eth_mask;
ret = mlx5_flow_item_acceptable(item, (const uint8_t *)mask,
* Item specification.
* @param[in] item_flags
* Bit-fields that holds the items detected until now.
+ * @param[in] dev
+ * Ethernet device flow is being created on.
* @param[out] error
* Pointer to error structure.
*
*/
int
mlx5_flow_validate_item_vlan(const struct rte_flow_item *item,
- int64_t item_flags,
+ uint64_t item_flags,
+ struct rte_eth_dev *dev,
struct rte_flow_error *error)
{
const struct rte_flow_item_vlan *spec = item->spec;
const struct rte_flow_item_vlan *mask = item->mask;
const struct rte_flow_item_vlan nic_mask = {
- .tci = RTE_BE16(0x0fff),
- .inner_type = RTE_BE16(0xffff),
+ .tci = RTE_BE16(UINT16_MAX),
+ .inner_type = RTE_BE16(UINT16_MAX),
};
uint16_t vlan_tag = 0;
const int tunnel = !!(item_flags & MLX5_FLOW_LAYER_TUNNEL);
int ret;
- const uint32_t l34m = tunnel ? (MLX5_FLOW_LAYER_INNER_L3 |
+ const uint64_t l34m = tunnel ? (MLX5_FLOW_LAYER_INNER_L3 |
MLX5_FLOW_LAYER_INNER_L4) :
(MLX5_FLOW_LAYER_OUTER_L3 |
MLX5_FLOW_LAYER_OUTER_L4);
- const uint32_t vlanm = tunnel ? MLX5_FLOW_LAYER_INNER_VLAN :
+ const uint64_t vlanm = tunnel ? MLX5_FLOW_LAYER_INNER_VLAN :
MLX5_FLOW_LAYER_OUTER_VLAN;
if (item_flags & vlanm)
return rte_flow_error_set(error, EINVAL,
RTE_FLOW_ERROR_TYPE_ITEM, item,
- "VLAN layer already configured");
+ "multiple VLAN layers not supported");
else if ((item_flags & l34m) != 0)
return rte_flow_error_set(error, EINVAL,
RTE_FLOW_ERROR_TYPE_ITEM, item,
error);
if (ret)
return ret;
+ if (!tunnel && mask->tci != RTE_BE16(0x0fff)) {
+ struct mlx5_priv *priv = dev->data->dev_private;
+
+ if (priv->vmwa_context) {
+ /*
+ * Non-NULL context means we have a virtual machine
+ * and SR-IOV enabled, we have to create VLAN interface
+ * to make hypervisor to setup E-Switch vport
+ * context correctly. We avoid creating the multiple
+ * VLAN interfaces, so we cannot support VLAN tag mask.
+ */
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM,
+ item,
+ "VLAN tag mask is not"
+ " supported in virtual"
+ " environment");
+ }
+ }
if (spec) {
vlan_tag = spec->tci;
vlan_tag &= mask->tci;
* Item specification.
* @param[in] item_flags
* Bit-fields that holds the items detected until now.
+ * @param[in] acc_mask
+ * Acceptable mask, if NULL default internal default mask
+ * will be used to check whether item fields are supported.
* @param[out] error
* Pointer to error structure.
*
*/
int
mlx5_flow_validate_item_ipv4(const struct rte_flow_item *item,
- int64_t item_flags,
+ uint64_t item_flags,
+ const struct rte_flow_item_ipv4 *acc_mask,
struct rte_flow_error *error)
{
const struct rte_flow_item_ipv4 *mask = item->mask;
+ const struct rte_flow_item_ipv4 *spec = item->spec;
const struct rte_flow_item_ipv4 nic_mask = {
.hdr = {
.src_addr = RTE_BE32(0xffffffff),
},
};
const int tunnel = !!(item_flags & MLX5_FLOW_LAYER_TUNNEL);
+ const uint64_t l3m = tunnel ? MLX5_FLOW_LAYER_INNER_L3 :
+ MLX5_FLOW_LAYER_OUTER_L3;
+ const uint64_t l4m = tunnel ? MLX5_FLOW_LAYER_INNER_L4 :
+ MLX5_FLOW_LAYER_OUTER_L4;
int ret;
+ uint8_t next_proto = 0xFF;
- if (item_flags & (tunnel ? MLX5_FLOW_LAYER_INNER_L3 :
- MLX5_FLOW_LAYER_OUTER_L3))
+ if (item_flags & MLX5_FLOW_LAYER_IPIP) {
+ if (mask && spec)
+ next_proto = mask->hdr.next_proto_id &
+ spec->hdr.next_proto_id;
+ if (next_proto == IPPROTO_IPIP || next_proto == IPPROTO_IPV6)
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM,
+ item,
+ "multiple tunnel "
+ "not supported");
+ }
+ if (item_flags & MLX5_FLOW_LAYER_IPV6_ENCAP)
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "wrong tunnel type - IPv6 specified "
+ "but IPv4 item provided");
+ if (item_flags & l3m)
return rte_flow_error_set(error, ENOTSUP,
RTE_FLOW_ERROR_TYPE_ITEM, item,
"multiple L3 layers not supported");
- else if (item_flags & (tunnel ? MLX5_FLOW_LAYER_INNER_L4 :
- MLX5_FLOW_LAYER_OUTER_L4))
+ else if (item_flags & l4m)
return rte_flow_error_set(error, EINVAL,
RTE_FLOW_ERROR_TYPE_ITEM, item,
"L3 cannot follow an L4 layer.");
+ else if ((item_flags & MLX5_FLOW_LAYER_NVGRE) &&
+ !(item_flags & MLX5_FLOW_LAYER_INNER_L2))
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "L3 cannot follow an NVGRE layer.");
if (!mask)
mask = &rte_flow_item_ipv4_mask;
+ else if (mask->hdr.next_proto_id != 0 &&
+ mask->hdr.next_proto_id != 0xff)
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM_MASK, mask,
+ "partial mask is not supported"
+ " for protocol");
ret = mlx5_flow_item_acceptable(item, (const uint8_t *)mask,
- (const uint8_t *)&nic_mask,
+ acc_mask ? (const uint8_t *)acc_mask
+ : (const uint8_t *)&nic_mask,
sizeof(struct rte_flow_item_ipv4),
error);
if (ret < 0)
* Item specification.
* @param[in] item_flags
* Bit-fields that holds the items detected until now.
+ * @param[in] acc_mask
+ * Acceptable mask, if NULL default internal default mask
+ * will be used to check whether item fields are supported.
* @param[out] error
* Pointer to error structure.
*
int
mlx5_flow_validate_item_ipv6(const struct rte_flow_item *item,
uint64_t item_flags,
+ const struct rte_flow_item_ipv6 *acc_mask,
struct rte_flow_error *error)
{
const struct rte_flow_item_ipv6 *mask = item->mask;
+ const struct rte_flow_item_ipv6 *spec = item->spec;
const struct rte_flow_item_ipv6 nic_mask = {
.hdr = {
.src_addr =
},
};
const int tunnel = !!(item_flags & MLX5_FLOW_LAYER_TUNNEL);
+ const uint64_t l3m = tunnel ? MLX5_FLOW_LAYER_INNER_L3 :
+ MLX5_FLOW_LAYER_OUTER_L3;
+ const uint64_t l4m = tunnel ? MLX5_FLOW_LAYER_INNER_L4 :
+ MLX5_FLOW_LAYER_OUTER_L4;
int ret;
+ uint8_t next_proto = 0xFF;
- if (item_flags & (tunnel ? MLX5_FLOW_LAYER_INNER_L3 :
- MLX5_FLOW_LAYER_OUTER_L3))
+ if (item_flags & MLX5_FLOW_LAYER_IPV6_ENCAP) {
+ if (mask && spec)
+ next_proto = mask->hdr.proto & spec->hdr.proto;
+ if (next_proto == IPPROTO_IPIP || next_proto == IPPROTO_IPV6)
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM,
+ item,
+ "multiple tunnel "
+ "not supported");
+ }
+ if (item_flags & MLX5_FLOW_LAYER_IPIP)
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "wrong tunnel type - IPv4 specified "
+ "but IPv6 item provided");
+ if (item_flags & l3m)
return rte_flow_error_set(error, ENOTSUP,
RTE_FLOW_ERROR_TYPE_ITEM, item,
"multiple L3 layers not supported");
- else if (item_flags & (tunnel ? MLX5_FLOW_LAYER_INNER_L4 :
- MLX5_FLOW_LAYER_OUTER_L4))
+ else if (item_flags & l4m)
return rte_flow_error_set(error, EINVAL,
RTE_FLOW_ERROR_TYPE_ITEM, item,
"L3 cannot follow an L4 layer.");
- /*
- * IPv6 is not recognised by the NIC inside a GRE tunnel.
- * Such support has to be disabled as the rule will be
- * accepted. Issue reproduced with Mellanox OFED 4.3-3.0.2.1 and
- * Mellanox OFED 4.4-1.0.0.0.
- */
- if (tunnel && item_flags & MLX5_FLOW_LAYER_GRE)
- return rte_flow_error_set(error, ENOTSUP,
+ else if ((item_flags & MLX5_FLOW_LAYER_NVGRE) &&
+ !(item_flags & MLX5_FLOW_LAYER_INNER_L2))
+ return rte_flow_error_set(error, EINVAL,
RTE_FLOW_ERROR_TYPE_ITEM, item,
- "IPv6 inside a GRE tunnel is"
- " not recognised.");
+ "L3 cannot follow an NVGRE layer.");
if (!mask)
mask = &rte_flow_item_ipv6_mask;
ret = mlx5_flow_item_acceptable(item, (const uint8_t *)mask,
- (const uint8_t *)&nic_mask,
+ acc_mask ? (const uint8_t *)acc_mask
+ : (const uint8_t *)&nic_mask,
sizeof(struct rte_flow_item_ipv6),
error);
if (ret < 0)
* @param[in] target_protocol
* The next protocol in the previous item.
* @param[in] flow_mask
- * mlx5 flow-specific (TCF, DV, verbs, etc.) supported header fields mask.
+ * mlx5 flow-specific (DV, verbs, etc.) supported header fields mask.
* @param[out] error
* Pointer to error structure.
*
{
const struct rte_flow_item_udp *mask = item->mask;
const int tunnel = !!(item_flags & MLX5_FLOW_LAYER_TUNNEL);
+ const uint64_t l3m = tunnel ? MLX5_FLOW_LAYER_INNER_L3 :
+ MLX5_FLOW_LAYER_OUTER_L3;
+ const uint64_t l4m = tunnel ? MLX5_FLOW_LAYER_INNER_L4 :
+ MLX5_FLOW_LAYER_OUTER_L4;
int ret;
if (target_protocol != 0xff && target_protocol != IPPROTO_UDP)
RTE_FLOW_ERROR_TYPE_ITEM, item,
"protocol filtering not compatible"
" with UDP layer");
- if (!(item_flags & (tunnel ? MLX5_FLOW_LAYER_INNER_L3 :
- MLX5_FLOW_LAYER_OUTER_L3)))
+ if (!(item_flags & l3m))
return rte_flow_error_set(error, EINVAL,
RTE_FLOW_ERROR_TYPE_ITEM, item,
"L3 is mandatory to filter on L4");
- if (item_flags & (tunnel ? MLX5_FLOW_LAYER_INNER_L4 :
- MLX5_FLOW_LAYER_OUTER_L4))
+ if (item_flags & l4m)
return rte_flow_error_set(error, EINVAL,
RTE_FLOW_ERROR_TYPE_ITEM, item,
- "L4 layer is already present");
+ "multiple L4 layers not supported");
if (!mask)
mask = &rte_flow_item_udp_mask;
ret = mlx5_flow_item_acceptable
{
const struct rte_flow_item_tcp *mask = item->mask;
const int tunnel = !!(item_flags & MLX5_FLOW_LAYER_TUNNEL);
+ const uint64_t l3m = tunnel ? MLX5_FLOW_LAYER_INNER_L3 :
+ MLX5_FLOW_LAYER_OUTER_L3;
+ const uint64_t l4m = tunnel ? MLX5_FLOW_LAYER_INNER_L4 :
+ MLX5_FLOW_LAYER_OUTER_L4;
int ret;
assert(flow_mask);
RTE_FLOW_ERROR_TYPE_ITEM, item,
"protocol filtering not compatible"
" with TCP layer");
- if (!(item_flags & (tunnel ? MLX5_FLOW_LAYER_INNER_L3 :
- MLX5_FLOW_LAYER_OUTER_L3)))
+ if (!(item_flags & l3m))
return rte_flow_error_set(error, EINVAL,
RTE_FLOW_ERROR_TYPE_ITEM, item,
"L3 is mandatory to filter on L4");
- if (item_flags & (tunnel ? MLX5_FLOW_LAYER_INNER_L4 :
- MLX5_FLOW_LAYER_OUTER_L4))
+ if (item_flags & l4m)
return rte_flow_error_set(error, EINVAL,
RTE_FLOW_ERROR_TYPE_ITEM, item,
- "L4 layer is already present");
+ "multiple L4 layers not supported");
if (!mask)
mask = &rte_flow_item_tcp_mask;
ret = mlx5_flow_item_acceptable
if (item_flags & MLX5_FLOW_LAYER_TUNNEL)
return rte_flow_error_set(error, ENOTSUP,
RTE_FLOW_ERROR_TYPE_ITEM, item,
- "a tunnel is already present");
+ "multiple tunnel layers not"
+ " supported");
/*
* Verify only UDPv4 is present as defined in
* https://tools.ietf.org/html/rfc7348
struct rte_eth_dev *dev,
struct rte_flow_error *error)
{
- struct priv *priv = dev->data->dev_private;
+ struct mlx5_priv *priv = dev->data->dev_private;
const struct rte_flow_item_vxlan_gpe *spec = item->spec;
const struct rte_flow_item_vxlan_gpe *mask = item->mask;
int ret;
if (item_flags & MLX5_FLOW_LAYER_TUNNEL)
return rte_flow_error_set(error, ENOTSUP,
RTE_FLOW_ERROR_TYPE_ITEM, item,
- "a tunnel is already present");
+ "multiple tunnel layers not"
+ " supported");
/*
* Verify only UDPv4 is present as defined in
* https://tools.ietf.org/html/rfc7348
" defined");
return 0;
}
+/**
+ * Validate GRE Key item.
+ *
+ * @param[in] item
+ * Item specification.
+ * @param[in] item_flags
+ * Bit flags to mark detected items.
+ * @param[in] gre_item
+ * Pointer to gre_item
+ * @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_item_gre_key(const struct rte_flow_item *item,
+ uint64_t item_flags,
+ const struct rte_flow_item *gre_item,
+ struct rte_flow_error *error)
+{
+ const rte_be32_t *mask = item->mask;
+ int ret = 0;
+ rte_be32_t gre_key_default_mask = RTE_BE32(UINT32_MAX);
+ const struct rte_flow_item_gre *gre_spec = gre_item->spec;
+ const struct rte_flow_item_gre *gre_mask = gre_item->mask;
+
+ if (item_flags & MLX5_FLOW_LAYER_GRE_KEY)
+ return rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "Multiple GRE key not support");
+ if (!(item_flags & MLX5_FLOW_LAYER_GRE))
+ return rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "No preceding GRE header");
+ if (item_flags & MLX5_FLOW_LAYER_INNER)
+ return rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "GRE key following a wrong item");
+ if (!gre_mask)
+ gre_mask = &rte_flow_item_gre_mask;
+ if (gre_spec && (gre_mask->c_rsvd0_ver & RTE_BE16(0x2000)) &&
+ !(gre_spec->c_rsvd0_ver & RTE_BE16(0x2000)))
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "Key bit must be on");
+
+ if (!mask)
+ mask = &gre_key_default_mask;
+ ret = mlx5_flow_item_acceptable
+ (item, (const uint8_t *)mask,
+ (const uint8_t *)&gre_key_default_mask,
+ sizeof(rte_be32_t), error);
+ return ret;
+}
/**
* Validate GRE item.
const struct rte_flow_item_gre *spec __rte_unused = item->spec;
const struct rte_flow_item_gre *mask = item->mask;
int ret;
+ const struct rte_flow_item_gre nic_mask = {
+ .c_rsvd0_ver = RTE_BE16(0xB000),
+ .protocol = RTE_BE16(UINT16_MAX),
+ };
if (target_protocol != 0xff && target_protocol != IPPROTO_GRE)
return rte_flow_error_set(error, EINVAL,
if (item_flags & MLX5_FLOW_LAYER_TUNNEL)
return rte_flow_error_set(error, ENOTSUP,
RTE_FLOW_ERROR_TYPE_ITEM, item,
- "a tunnel is already present");
+ "multiple tunnel layers not"
+ " supported");
if (!(item_flags & MLX5_FLOW_LAYER_OUTER_L3))
return rte_flow_error_set(error, ENOTSUP,
RTE_FLOW_ERROR_TYPE_ITEM, item,
mask = &rte_flow_item_gre_mask;
ret = mlx5_flow_item_acceptable
(item, (const uint8_t *)mask,
- (const uint8_t *)&rte_flow_item_gre_mask,
+ (const uint8_t *)&nic_mask,
sizeof(struct rte_flow_item_gre), error);
if (ret < 0)
return ret;
+#ifndef HAVE_MLX5DV_DR
#ifndef HAVE_IBV_DEVICE_MPLS_SUPPORT
if (spec && (spec->protocol & mask->protocol))
return rte_flow_error_set(error, ENOTSUP,
"without MPLS support the"
" specification cannot be used for"
" filtering");
+#endif
#endif
return 0;
}
/**
* Validate MPLS item.
*
+ * @param[in] dev
+ * Pointer to the rte_eth_dev structure.
* @param[in] item
* Item specification.
* @param[in] item_flags
* Bit-fields that holds the items detected until now.
- * @param[in] target_protocol
- * The next protocol in the previous item.
+ * @param[in] prev_layer
+ * The protocol layer indicated in previous item.
* @param[out] error
* Pointer to error structure.
*
* 0 on success, a negative errno value otherwise and rte_errno is set.
*/
int
-mlx5_flow_validate_item_mpls(const struct rte_flow_item *item __rte_unused,
+mlx5_flow_validate_item_mpls(struct rte_eth_dev *dev __rte_unused,
+ const struct rte_flow_item *item __rte_unused,
uint64_t item_flags __rte_unused,
- uint8_t target_protocol __rte_unused,
+ uint64_t prev_layer __rte_unused,
struct rte_flow_error *error)
{
#ifdef HAVE_IBV_DEVICE_MPLS_SUPPORT
const struct rte_flow_item_mpls *mask = item->mask;
+ struct mlx5_priv *priv = dev->data->dev_private;
int ret;
- if (target_protocol != 0xff && target_protocol != IPPROTO_MPLS)
+ if (!priv->config.mpls_en)
+ return rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "MPLS not supported or"
+ " disabled in firmware"
+ " configuration.");
+ /* MPLS over IP, UDP, GRE is allowed */
+ if (!(prev_layer & (MLX5_FLOW_LAYER_OUTER_L3 |
+ MLX5_FLOW_LAYER_OUTER_L4_UDP |
+ MLX5_FLOW_LAYER_GRE)))
return rte_flow_error_set(error, EINVAL,
RTE_FLOW_ERROR_TYPE_ITEM, item,
"protocol filtering not compatible"
" with MPLS layer");
- if (item_flags & MLX5_FLOW_LAYER_TUNNEL)
+ /* Multi-tunnel isn't allowed but MPLS over GRE is an exception. */
+ if ((item_flags & MLX5_FLOW_LAYER_TUNNEL) &&
+ !(item_flags & MLX5_FLOW_LAYER_GRE))
return rte_flow_error_set(error, ENOTSUP,
RTE_FLOW_ERROR_TYPE_ITEM, item,
- "a tunnel is already"
- " present");
+ "multiple tunnel layers not"
+ " supported");
if (!mask)
mask = &rte_flow_item_mpls_mask;
ret = mlx5_flow_item_acceptable
" update.");
}
+/**
+ * Validate NVGRE item.
+ *
+ * @param[in] item
+ * Item specification.
+ * @param[in] item_flags
+ * Bit flags to mark detected items.
+ * @param[in] target_protocol
+ * The next protocol in the previous item.
+ * @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_item_nvgre(const struct rte_flow_item *item,
+ uint64_t item_flags,
+ uint8_t target_protocol,
+ struct rte_flow_error *error)
+{
+ const struct rte_flow_item_nvgre *mask = item->mask;
+ int ret;
+
+ if (target_protocol != 0xff && target_protocol != IPPROTO_GRE)
+ return rte_flow_error_set(error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "protocol filtering not compatible"
+ " with this GRE layer");
+ if (item_flags & MLX5_FLOW_LAYER_TUNNEL)
+ return rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "multiple tunnel layers not"
+ " supported");
+ if (!(item_flags & MLX5_FLOW_LAYER_OUTER_L3))
+ return rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_ITEM, item,
+ "L3 Layer is missing");
+ if (!mask)
+ mask = &rte_flow_item_nvgre_mask;
+ ret = mlx5_flow_item_acceptable
+ (item, (const uint8_t *)mask,
+ (const uint8_t *)&rte_flow_item_nvgre_mask,
+ sizeof(struct rte_flow_item_nvgre), error);
+ if (ret < 0)
+ return ret;
+ return 0;
+}
+
static int
flow_null_validate(struct rte_eth_dev *dev __rte_unused,
const struct rte_flow_attr *attr __rte_unused,
const struct rte_flow_item items[] __rte_unused,
const struct rte_flow_action actions[] __rte_unused,
- struct rte_flow_error *error __rte_unused)
+ bool external __rte_unused,
+ struct rte_flow_error *error)
{
- rte_errno = ENOTSUP;
- return -rte_errno;
+ return rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, NULL);
}
static struct mlx5_flow *
flow_null_prepare(const struct rte_flow_attr *attr __rte_unused,
const struct rte_flow_item items[] __rte_unused,
const struct rte_flow_action actions[] __rte_unused,
- uint64_t *item_flags __rte_unused,
- uint64_t *action_flags __rte_unused,
- struct rte_flow_error *error __rte_unused)
+ struct rte_flow_error *error)
{
- rte_errno = ENOTSUP;
+ rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, NULL);
return NULL;
}
const struct rte_flow_attr *attr __rte_unused,
const struct rte_flow_item items[] __rte_unused,
const struct rte_flow_action actions[] __rte_unused,
- struct rte_flow_error *error __rte_unused)
+ struct rte_flow_error *error)
{
- rte_errno = ENOTSUP;
- return -rte_errno;
+ return rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, NULL);
}
static int
flow_null_apply(struct rte_eth_dev *dev __rte_unused,
struct rte_flow *flow __rte_unused,
- struct rte_flow_error *error __rte_unused)
+ struct rte_flow_error *error)
{
- rte_errno = ENOTSUP;
- return -rte_errno;
+ return rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, NULL);
}
static void
{
}
+static int
+flow_null_query(struct rte_eth_dev *dev __rte_unused,
+ struct rte_flow *flow __rte_unused,
+ const struct rte_flow_action *actions __rte_unused,
+ void *data __rte_unused,
+ struct rte_flow_error *error)
+{
+ return rte_flow_error_set(error, ENOTSUP,
+ RTE_FLOW_ERROR_TYPE_UNSPECIFIED, NULL, NULL);
+}
+
/* Void driver to protect from null pointer reference. */
const struct mlx5_flow_driver_ops mlx5_flow_null_drv_ops = {
.validate = flow_null_validate,
.apply = flow_null_apply,
.remove = flow_null_remove,
.destroy = flow_null_destroy,
+ .query = flow_null_query,
};
/**
* Pointer to the flow attributes.
*
* @return
- * flow driver type if supported, MLX5_FLOW_TYPE_MAX otherwise.
+ * flow driver type, MLX5_FLOW_TYPE_MAX otherwise.
*/
static enum mlx5_flow_drv_type
-flow_get_drv_type(struct rte_eth_dev *dev __rte_unused,
- const struct rte_flow_attr *attr)
+flow_get_drv_type(struct rte_eth_dev *dev, const struct rte_flow_attr *attr)
{
- struct priv *priv __rte_unused = dev->data->dev_private;
+ struct mlx5_priv *priv = dev->data->dev_private;
enum mlx5_flow_drv_type type = MLX5_FLOW_TYPE_MAX;
- if (attr->transfer) {
- type = MLX5_FLOW_TYPE_TCF;
- } else {
-#ifdef HAVE_IBV_FLOW_DV_SUPPORT
- type = priv->config.dv_flow_en ? MLX5_FLOW_TYPE_DV :
- MLX5_FLOW_TYPE_VERBS;
-#else
- type = MLX5_FLOW_TYPE_VERBS;
-#endif
- }
+ if (attr->transfer && priv->config.dv_esw_en)
+ type = MLX5_FLOW_TYPE_DV;
+ if (!attr->transfer)
+ type = priv->config.dv_flow_en ? MLX5_FLOW_TYPE_DV :
+ MLX5_FLOW_TYPE_VERBS;
return type;
}
* Pointer to the list of items.
* @param[in] actions
* Pointer to the list of actions.
+ * @param[in] external
+ * This flow rule is created by request external to PMD.
* @param[out] error
* Pointer to the error structure.
*
* @return
- * 0 on success, a negative errno value otherwise and rte_ernno is set.
+ * 0 on success, a negative errno value otherwise and rte_errno is set.
*/
static inline int
flow_drv_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)
+ bool external, 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, error);
+ return fops->validate(dev, attr, items, actions, external, error);
}
/**
* calculates the size of memory required for device flow, allocates the memory,
* initializes the device flow and returns the pointer.
*
+ * @note
+ * This function initializes device flow structure such as dv or verbs in
+ * struct mlx5_flow. However, it is caller's responsibility to initialize the
+ * rest. For example, adding returning device flow to flow->dev_flow list and
+ * setting backward reference to the flow should be done out of this function.
+ * layers field is not filled either.
+ *
* @param[in] attr
* Pointer to the flow attributes.
* @param[in] items
* Pointer to the list of items.
* @param[in] actions
* Pointer to the list of actions.
- * @param[out] item_flags
- * Pointer to bit mask of all items detected.
- * @param[out] action_flags
- * Pointer to bit mask of all actions detected.
* @param[out] error
* Pointer to the error structure.
*
* @return
- * Pointer to device flow on success, otherwise NULL and rte_ernno is set.
+ * Pointer to device flow on success, otherwise NULL and rte_errno is set.
*/
static inline struct mlx5_flow *
-flow_drv_prepare(struct rte_flow *flow,
+flow_drv_prepare(const struct rte_flow *flow,
const struct rte_flow_attr *attr,
const struct rte_flow_item items[],
const struct rte_flow_action actions[],
- uint64_t *item_flags,
- uint64_t *action_flags,
struct rte_flow_error *error)
{
const struct mlx5_flow_driver_ops *fops;
assert(type > MLX5_FLOW_TYPE_MIN && type < MLX5_FLOW_TYPE_MAX);
fops = flow_get_drv_ops(type);
- return fops->prepare(attr, items, actions, item_flags, action_flags,
- error);
+ return fops->prepare(attr, items, actions, error);
}
/**
* translates a generic flow into a driver flow. flow_drv_prepare() must
* precede.
*
+ * @note
+ * dev_flow->layers could be filled as a result of parsing during translation
+ * if needed by flow_drv_apply(). dev_flow->flow->actions can also be filled
+ * if necessary. As a flow can have multiple dev_flows by RSS flow expansion,
+ * flow->actions could be overwritten even though all the expanded dev_flows
+ * have the same actions.
*
* @param[in] dev
* Pointer to the rte dev structure.
* Pointer to the error structure.
*
* @return
- * 0 on success, a negative errno value otherwise and rte_ernno is set.
+ * 0 on success, a negative errno value otherwise and rte_errno is set.
*/
static inline int
flow_drv_translate(struct rte_eth_dev *dev, struct mlx5_flow *dev_flow,
* Flow driver remove API. This abstracts calling driver specific functions.
* Parent flow (rte_flow) should have driver type (drv_type). It removes a flow
* on device. All the resources of the flow should be freed by calling
- * flow_dv_destroy().
+ * flow_drv_destroy().
*
* @param[in] dev
* Pointer to Ethernet device.
{
int ret;
- ret = flow_drv_validate(dev, attr, items, actions, error);
+ ret = flow_drv_validate(dev, attr, items, actions, true, error);
if (ret < 0)
return ret;
return 0;
* Pointer to the RSS action if exist, else return NULL.
*/
static const struct rte_flow_action_rss*
-mlx5_flow_get_rss_action(const struct rte_flow_action actions[])
+flow_get_rss_action(const struct rte_flow_action actions[])
{
for (; actions->type != RTE_FLOW_ACTION_TYPE_END; actions++) {
switch (actions->type) {
}
static unsigned int
-mlx5_find_graph_root(const struct rte_flow_item pattern[], uint32_t rss_level)
+find_graph_root(const struct rte_flow_item pattern[], uint32_t rss_level)
{
const struct rte_flow_item *item;
unsigned int has_vlan = 0;
* Pattern specification (list terminated by the END pattern item).
* @param[in] actions
* Associated actions (list terminated by the END action).
+ * @param[in] external
+ * This flow rule is created by request external to PMD.
* @param[out] error
* Perform verbose error reporting if not NULL.
*
* A flow on success, NULL otherwise and rte_errno is set.
*/
static struct rte_flow *
-mlx5_flow_list_create(struct rte_eth_dev *dev,
- struct mlx5_flows *list,
- const struct rte_flow_attr *attr,
- const struct rte_flow_item items[],
- const struct rte_flow_action actions[],
- struct rte_flow_error *error)
+flow_list_create(struct rte_eth_dev *dev, struct mlx5_flows *list,
+ 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)
{
struct rte_flow *flow = NULL;
struct mlx5_flow *dev_flow;
- uint64_t action_flags = 0;
- uint64_t item_flags = 0;
const struct rte_flow_action_rss *rss;
union {
struct rte_flow_expand_rss buf;
uint32_t i;
uint32_t flow_size;
- ret = flow_drv_validate(dev, attr, items, actions, error);
+ ret = flow_drv_validate(dev, attr, items, actions, external, error);
if (ret < 0)
return NULL;
flow_size = sizeof(struct rte_flow);
- rss = mlx5_flow_get_rss_action(actions);
+ rss = flow_get_rss_action(actions);
if (rss)
flow_size += RTE_ALIGN_CEIL(rss->queue_num * sizeof(uint16_t),
sizeof(void *));
else
flow_size += RTE_ALIGN_CEIL(sizeof(uint16_t), sizeof(void *));
flow = rte_calloc(__func__, 1, flow_size, 0);
+ if (!flow) {
+ rte_errno = ENOMEM;
+ return NULL;
+ }
flow->drv_type = flow_get_drv_type(dev, attr);
+ flow->ingress = attr->ingress;
+ flow->transfer = attr->transfer;
assert(flow->drv_type > MLX5_FLOW_TYPE_MIN &&
flow->drv_type < MLX5_FLOW_TYPE_MAX);
flow->queue = (void *)(flow + 1);
if (rss && rss->types) {
unsigned int graph_root;
- graph_root = mlx5_find_graph_root(items, rss->level);
+ graph_root = find_graph_root(items, rss->level);
ret = rte_flow_expand_rss(buf, sizeof(expand_buffer.buffer),
items, rss->types,
mlx5_support_expansion,
}
for (i = 0; i < buf->entries; ++i) {
dev_flow = flow_drv_prepare(flow, attr, buf->entry[i].pattern,
- actions, &item_flags, &action_flags,
- error);
+ actions, error);
if (!dev_flow)
goto error;
dev_flow->flow = flow;
+ dev_flow->external = external;
LIST_INSERT_HEAD(&flow->dev_flows, dev_flow, next);
ret = flow_drv_translate(dev, dev_flow, attr,
buf->entry[i].pattern,
goto error;
}
TAILQ_INSERT_TAIL(list, flow, next);
- mlx5_flow_rxq_flags_set(dev, flow);
+ flow_rxq_flags_set(dev, flow);
return flow;
error:
ret = rte_errno; /* Save rte_errno before cleanup. */
return NULL;
}
+/**
+ * Create a dedicated flow rule on e-switch table 0 (root table), to direct all
+ * incoming packets to table 1.
+ *
+ * Other flow rules, requested for group n, will be created in
+ * e-switch table n+1.
+ * Jump action to e-switch group n will be created to group n+1.
+ *
+ * Used when working in switchdev mode, to utilise advantages of table 1
+ * and above.
+ *
+ * @param dev
+ * Pointer to Ethernet device.
+ *
+ * @return
+ * Pointer to flow on success, NULL otherwise and rte_errno is set.
+ */
+struct rte_flow *
+mlx5_flow_create_esw_table_zero_flow(struct rte_eth_dev *dev)
+{
+ const struct rte_flow_attr attr = {
+ .group = 0,
+ .priority = 0,
+ .ingress = 1,
+ .egress = 0,
+ .transfer = 1,
+ };
+ const struct rte_flow_item pattern = {
+ .type = RTE_FLOW_ITEM_TYPE_END,
+ };
+ struct rte_flow_action_jump jump = {
+ .group = 1,
+ };
+ const struct rte_flow_action actions[] = {
+ {
+ .type = RTE_FLOW_ACTION_TYPE_JUMP,
+ .conf = &jump,
+ },
+ {
+ .type = RTE_FLOW_ACTION_TYPE_END,
+ },
+ };
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct rte_flow_error error;
+
+ return flow_list_create(dev, &priv->ctrl_flows, &attr, &pattern,
+ actions, false, &error);
+}
+
/**
* Create a flow.
*
const struct rte_flow_action actions[],
struct rte_flow_error *error)
{
- return mlx5_flow_list_create
- (dev, &((struct priv *)dev->data->dev_private)->flows,
- attr, items, actions, error);
+ struct mlx5_priv *priv = dev->data->dev_private;
+
+ return flow_list_create(dev, &priv->flows,
+ attr, items, actions, true, error);
}
/**
* Flow to destroy.
*/
static void
-mlx5_flow_list_destroy(struct rte_eth_dev *dev, struct mlx5_flows *list,
- struct rte_flow *flow)
+flow_list_destroy(struct rte_eth_dev *dev, struct mlx5_flows *list,
+ struct rte_flow *flow)
{
- flow_drv_destroy(dev, flow);
- TAILQ_REMOVE(list, flow, next);
/*
* Update RX queue flags only if port is started, otherwise it is
* already clean.
*/
if (dev->data->dev_started)
- mlx5_flow_rxq_flags_trim(dev, flow);
+ flow_rxq_flags_trim(dev, flow);
+ flow_drv_destroy(dev, flow);
+ TAILQ_REMOVE(list, flow, next);
+ rte_free(flow->fdir);
rte_free(flow);
}
struct rte_flow *flow;
flow = TAILQ_FIRST(list);
- mlx5_flow_list_destroy(dev, list, flow);
+ flow_list_destroy(dev, list, flow);
}
}
TAILQ_FOREACH_REVERSE(flow, list, mlx5_flows, next)
flow_drv_remove(dev, flow);
- mlx5_flow_rxq_flags_clear(dev);
+ flow_rxq_flags_clear(dev);
}
/**
ret = flow_drv_apply(dev, flow, &error);
if (ret < 0)
goto error;
- mlx5_flow_rxq_flags_set(dev, flow);
+ flow_rxq_flags_set(dev, flow);
}
return 0;
error:
int
mlx5_flow_verify(struct rte_eth_dev *dev)
{
- struct priv *priv = dev->data->dev_private;
+ struct mlx5_priv *priv = dev->data->dev_private;
struct rte_flow *flow;
int ret = 0;
struct rte_flow_item_vlan *vlan_spec,
struct rte_flow_item_vlan *vlan_mask)
{
- struct priv *priv = dev->data->dev_private;
+ struct mlx5_priv *priv = dev->data->dev_private;
const struct rte_flow_attr attr = {
.ingress = 1,
.priority = MLX5_FLOW_PRIO_RSVD,
struct rte_flow_error error;
unsigned int i;
- if (!priv->reta_idx_n) {
- rte_errno = EINVAL;
- return -rte_errno;
+ if (!priv->reta_idx_n || !priv->rxqs_n) {
+ return 0;
}
for (i = 0; i != priv->reta_idx_n; ++i)
queue[i] = (*priv->reta_idx)[i];
- flow = mlx5_flow_list_create(dev, &priv->ctrl_flows, &attr, items,
- actions, &error);
+ flow = flow_list_create(dev, &priv->ctrl_flows,
+ &attr, items, actions, false, &error);
if (!flow)
return -rte_errno;
return 0;
struct rte_flow *flow,
struct rte_flow_error *error __rte_unused)
{
- struct priv *priv = dev->data->dev_private;
+ struct mlx5_priv *priv = dev->data->dev_private;
- mlx5_flow_list_destroy(dev, &priv->flows, flow);
+ flow_list_destroy(dev, &priv->flows, flow);
return 0;
}
mlx5_flow_flush(struct rte_eth_dev *dev,
struct rte_flow_error *error __rte_unused)
{
- struct priv *priv = dev->data->dev_private;
+ struct mlx5_priv *priv = dev->data->dev_private;
mlx5_flow_list_flush(dev, &priv->flows);
return 0;
int enable,
struct rte_flow_error *error)
{
- struct priv *priv = dev->data->dev_private;
+ struct mlx5_priv *priv = dev->data->dev_private;
if (dev->data->dev_started) {
rte_flow_error_set(error, EBUSY,
}
/**
- * Query flow counter.
+ * Query a flow.
*
- * @param flow
- * Pointer to the flow.
- *
- * @return
- * 0 on success, a negative errno value otherwise and rte_errno is set.
+ * @see rte_flow_query()
+ * @see rte_flow_ops
*/
static int
-mlx5_flow_query_count(struct rte_flow *flow __rte_unused,
- void *data __rte_unused,
- struct rte_flow_error *error)
-{
-#ifdef HAVE_IBV_DEVICE_COUNTERS_SET_SUPPORT
- if (flow->actions & MLX5_FLOW_ACTION_COUNT) {
- struct rte_flow_query_count *qc = data;
- uint64_t counters[2] = {0, 0};
- struct ibv_query_counter_set_attr query_cs_attr = {
- .cs = flow->counter->cs,
- .query_flags = IBV_COUNTER_SET_FORCE_UPDATE,
- };
- struct ibv_counter_set_data query_out = {
- .out = counters,
- .outlen = 2 * sizeof(uint64_t),
- };
- int err = mlx5_glue->query_counter_set(&query_cs_attr,
- &query_out);
+flow_drv_query(struct rte_eth_dev *dev,
+ struct rte_flow *flow,
+ const struct rte_flow_action *actions,
+ void *data,
+ struct rte_flow_error *error)
+{
+ const struct mlx5_flow_driver_ops *fops;
+ enum mlx5_flow_drv_type ftype = flow->drv_type;
- if (err)
- return rte_flow_error_set
- (error, err,
- RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
- NULL,
- "cannot read counter");
- qc->hits_set = 1;
- qc->bytes_set = 1;
- qc->hits = counters[0] - flow->counter->hits;
- qc->bytes = counters[1] - flow->counter->bytes;
- if (qc->reset) {
- flow->counter->hits = counters[0];
- flow->counter->bytes = counters[1];
- }
- return 0;
- }
- return rte_flow_error_set(error, EINVAL,
- RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
- NULL,
- "flow does not have counter");
-#endif
- return rte_flow_error_set(error, ENOTSUP,
- RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
- NULL,
- "counters are not available");
+ assert(ftype > MLX5_FLOW_TYPE_MIN && ftype < MLX5_FLOW_TYPE_MAX);
+ fops = flow_get_drv_ops(ftype);
+
+ return fops->query(dev, flow, actions, data, error);
}
/**
- * Query a flows.
+ * Query a flow.
*
* @see rte_flow_query()
* @see rte_flow_ops
*/
int
-mlx5_flow_query(struct rte_eth_dev *dev __rte_unused,
+mlx5_flow_query(struct rte_eth_dev *dev,
struct rte_flow *flow,
const struct rte_flow_action *actions,
void *data,
struct rte_flow_error *error)
{
- int ret = 0;
+ int ret;
- for (; actions->type != RTE_FLOW_ACTION_TYPE_END; actions++) {
- switch (actions->type) {
- case RTE_FLOW_ACTION_TYPE_VOID:
- break;
- case RTE_FLOW_ACTION_TYPE_COUNT:
- ret = mlx5_flow_query_count(flow, data, error);
- break;
- default:
- return rte_flow_error_set(error, ENOTSUP,
- RTE_FLOW_ERROR_TYPE_ACTION,
- actions,
- "action not supported");
- }
- if (ret < 0)
- return ret;
- }
+ ret = flow_drv_query(dev, flow, actions, data, error);
+ if (ret < 0)
+ return ret;
return 0;
}
* 0 on success, a negative errno value otherwise and rte_errno is set.
*/
static int
-mlx5_fdir_filter_convert(struct rte_eth_dev *dev,
+flow_fdir_filter_convert(struct rte_eth_dev *dev,
const struct rte_eth_fdir_filter *fdir_filter,
struct mlx5_fdir *attributes)
{
- struct priv *priv = dev->data->dev_private;
+ struct mlx5_priv *priv = dev->data->dev_private;
const struct rte_eth_fdir_input *input = &fdir_filter->input;
const struct rte_eth_fdir_masks *mask =
&dev->data->dev_conf.fdir_conf.mask;
case RTE_ETH_FLOW_NONFRAG_IPV4_UDP:
case RTE_ETH_FLOW_NONFRAG_IPV4_TCP:
case RTE_ETH_FLOW_NONFRAG_IPV4_OTHER:
- attributes->l3.ipv4.hdr = (struct ipv4_hdr){
+ attributes->l3.ipv4.hdr = (struct rte_ipv4_hdr){
.src_addr = input->flow.ip4_flow.src_ip,
.dst_addr = input->flow.ip4_flow.dst_ip,
.time_to_live = input->flow.ip4_flow.ttl,
.type_of_service = input->flow.ip4_flow.tos,
};
- attributes->l3_mask.ipv4.hdr = (struct ipv4_hdr){
+ attributes->l3_mask.ipv4.hdr = (struct rte_ipv4_hdr){
.src_addr = mask->ipv4_mask.src_ip,
.dst_addr = mask->ipv4_mask.dst_ip,
.time_to_live = mask->ipv4_mask.ttl,
case RTE_ETH_FLOW_NONFRAG_IPV6_UDP:
case RTE_ETH_FLOW_NONFRAG_IPV6_TCP:
case RTE_ETH_FLOW_NONFRAG_IPV6_OTHER:
- attributes->l3.ipv6.hdr = (struct ipv6_hdr){
+ attributes->l3.ipv6.hdr = (struct rte_ipv6_hdr){
.hop_limits = input->flow.ipv6_flow.hop_limits,
.proto = input->flow.ipv6_flow.proto,
};
/* Handle L4. */
switch (fdir_filter->input.flow_type) {
case RTE_ETH_FLOW_NONFRAG_IPV4_UDP:
- attributes->l4.udp.hdr = (struct udp_hdr){
+ attributes->l4.udp.hdr = (struct rte_udp_hdr){
.src_port = input->flow.udp4_flow.src_port,
.dst_port = input->flow.udp4_flow.dst_port,
};
- attributes->l4_mask.udp.hdr = (struct udp_hdr){
+ attributes->l4_mask.udp.hdr = (struct rte_udp_hdr){
.src_port = mask->src_port_mask,
.dst_port = mask->dst_port_mask,
};
};
break;
case RTE_ETH_FLOW_NONFRAG_IPV4_TCP:
- attributes->l4.tcp.hdr = (struct tcp_hdr){
+ attributes->l4.tcp.hdr = (struct rte_tcp_hdr){
.src_port = input->flow.tcp4_flow.src_port,
.dst_port = input->flow.tcp4_flow.dst_port,
};
- attributes->l4_mask.tcp.hdr = (struct tcp_hdr){
+ attributes->l4_mask.tcp.hdr = (struct rte_tcp_hdr){
.src_port = mask->src_port_mask,
.dst_port = mask->dst_port_mask,
};
};
break;
case RTE_ETH_FLOW_NONFRAG_IPV6_UDP:
- attributes->l4.udp.hdr = (struct udp_hdr){
+ attributes->l4.udp.hdr = (struct rte_udp_hdr){
.src_port = input->flow.udp6_flow.src_port,
.dst_port = input->flow.udp6_flow.dst_port,
};
- attributes->l4_mask.udp.hdr = (struct udp_hdr){
+ attributes->l4_mask.udp.hdr = (struct rte_udp_hdr){
.src_port = mask->src_port_mask,
.dst_port = mask->dst_port_mask,
};
};
break;
case RTE_ETH_FLOW_NONFRAG_IPV6_TCP:
- attributes->l4.tcp.hdr = (struct tcp_hdr){
+ attributes->l4.tcp.hdr = (struct rte_tcp_hdr){
.src_port = input->flow.tcp6_flow.src_port,
.dst_port = input->flow.tcp6_flow.dst_port,
};
- attributes->l4_mask.tcp.hdr = (struct tcp_hdr){
+ attributes->l4_mask.tcp.hdr = (struct rte_tcp_hdr){
.src_port = mask->src_port_mask,
.dst_port = mask->dst_port_mask,
};
return 0;
}
+#define FLOW_FDIR_CMP(f1, f2, fld) \
+ memcmp(&(f1)->fld, &(f2)->fld, sizeof(f1->fld))
+
+/**
+ * Compare two FDIR flows. If items and actions are identical, the two flows are
+ * regarded as same.
+ *
+ * @param dev
+ * Pointer to Ethernet device.
+ * @param f1
+ * FDIR flow to compare.
+ * @param f2
+ * FDIR flow to compare.
+ *
+ * @return
+ * Zero on match, 1 otherwise.
+ */
+static int
+flow_fdir_cmp(const struct mlx5_fdir *f1, const struct mlx5_fdir *f2)
+{
+ if (FLOW_FDIR_CMP(f1, f2, attr) ||
+ FLOW_FDIR_CMP(f1, f2, l2) ||
+ FLOW_FDIR_CMP(f1, f2, l2_mask) ||
+ FLOW_FDIR_CMP(f1, f2, l3) ||
+ FLOW_FDIR_CMP(f1, f2, l3_mask) ||
+ FLOW_FDIR_CMP(f1, f2, l4) ||
+ FLOW_FDIR_CMP(f1, f2, l4_mask) ||
+ FLOW_FDIR_CMP(f1, f2, actions[0].type))
+ return 1;
+ if (f1->actions[0].type == RTE_FLOW_ACTION_TYPE_QUEUE &&
+ FLOW_FDIR_CMP(f1, f2, queue))
+ return 1;
+ return 0;
+}
+
+/**
+ * Search device flow list to find out a matched FDIR flow.
+ *
+ * @param dev
+ * Pointer to Ethernet device.
+ * @param fdir_flow
+ * FDIR flow to lookup.
+ *
+ * @return
+ * Pointer of flow if found, NULL otherwise.
+ */
+static struct rte_flow *
+flow_fdir_filter_lookup(struct rte_eth_dev *dev, struct mlx5_fdir *fdir_flow)
+{
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct rte_flow *flow = NULL;
+
+ assert(fdir_flow);
+ TAILQ_FOREACH(flow, &priv->flows, next) {
+ if (flow->fdir && !flow_fdir_cmp(flow->fdir, fdir_flow)) {
+ DRV_LOG(DEBUG, "port %u found FDIR flow %p",
+ dev->data->port_id, (void *)flow);
+ break;
+ }
+ }
+ return flow;
+}
+
/**
* Add new flow director filter and store it in list.
*
* 0 on success, a negative errno value otherwise and rte_errno is set.
*/
static int
-mlx5_fdir_filter_add(struct rte_eth_dev *dev,
+flow_fdir_filter_add(struct rte_eth_dev *dev,
const struct rte_eth_fdir_filter *fdir_filter)
{
- struct priv *priv = dev->data->dev_private;
- struct mlx5_fdir attributes = {
- .attr.group = 0,
- .l2_mask = {
- .dst.addr_bytes = "\x00\x00\x00\x00\x00\x00",
- .src.addr_bytes = "\x00\x00\x00\x00\x00\x00",
- .type = 0,
- },
- };
- struct rte_flow_error error;
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct mlx5_fdir *fdir_flow;
struct rte_flow *flow;
int ret;
- ret = mlx5_fdir_filter_convert(dev, fdir_filter, &attributes);
+ fdir_flow = rte_zmalloc(__func__, sizeof(*fdir_flow), 0);
+ if (!fdir_flow) {
+ rte_errno = ENOMEM;
+ return -rte_errno;
+ }
+ ret = flow_fdir_filter_convert(dev, fdir_filter, fdir_flow);
if (ret)
- return ret;
- flow = mlx5_flow_list_create(dev, &priv->flows, &attributes.attr,
- attributes.items, attributes.actions,
- &error);
+ goto error;
+ flow = flow_fdir_filter_lookup(dev, fdir_flow);
if (flow) {
- DRV_LOG(DEBUG, "port %u FDIR created %p", dev->data->port_id,
- (void *)flow);
- return 0;
+ rte_errno = EEXIST;
+ goto error;
}
+ flow = flow_list_create(dev, &priv->flows, &fdir_flow->attr,
+ fdir_flow->items, fdir_flow->actions, true,
+ NULL);
+ if (!flow)
+ goto error;
+ assert(!flow->fdir);
+ flow->fdir = fdir_flow;
+ DRV_LOG(DEBUG, "port %u created FDIR flow %p",
+ dev->data->port_id, (void *)flow);
+ return 0;
+error:
+ rte_free(fdir_flow);
return -rte_errno;
}
* 0 on success, a negative errno value otherwise and rte_errno is set.
*/
static int
-mlx5_fdir_filter_delete(struct rte_eth_dev *dev __rte_unused,
- const struct rte_eth_fdir_filter *fdir_filter
- __rte_unused)
+flow_fdir_filter_delete(struct rte_eth_dev *dev,
+ const struct rte_eth_fdir_filter *fdir_filter)
{
- rte_errno = ENOTSUP;
- return -rte_errno;
+ struct mlx5_priv *priv = dev->data->dev_private;
+ struct rte_flow *flow;
+ struct mlx5_fdir fdir_flow = {
+ .attr.group = 0,
+ };
+ int ret;
+
+ ret = flow_fdir_filter_convert(dev, fdir_filter, &fdir_flow);
+ if (ret)
+ return -rte_errno;
+ flow = flow_fdir_filter_lookup(dev, &fdir_flow);
+ if (!flow) {
+ rte_errno = ENOENT;
+ return -rte_errno;
+ }
+ flow_list_destroy(dev, &priv->flows, flow);
+ DRV_LOG(DEBUG, "port %u deleted FDIR flow %p",
+ dev->data->port_id, (void *)flow);
+ return 0;
}
/**
* 0 on success, a negative errno value otherwise and rte_errno is set.
*/
static int
-mlx5_fdir_filter_update(struct rte_eth_dev *dev,
+flow_fdir_filter_update(struct rte_eth_dev *dev,
const struct rte_eth_fdir_filter *fdir_filter)
{
int ret;
- ret = mlx5_fdir_filter_delete(dev, fdir_filter);
+ ret = flow_fdir_filter_delete(dev, fdir_filter);
if (ret)
return ret;
- return mlx5_fdir_filter_add(dev, fdir_filter);
+ return flow_fdir_filter_add(dev, fdir_filter);
}
/**
* Pointer to Ethernet device.
*/
static void
-mlx5_fdir_filter_flush(struct rte_eth_dev *dev)
+flow_fdir_filter_flush(struct rte_eth_dev *dev)
{
- struct priv *priv = dev->data->dev_private;
+ struct mlx5_priv *priv = dev->data->dev_private;
mlx5_flow_list_flush(dev, &priv->flows);
}
* Resulting flow director information.
*/
static void
-mlx5_fdir_info_get(struct rte_eth_dev *dev, struct rte_eth_fdir_info *fdir_info)
+flow_fdir_info_get(struct rte_eth_dev *dev, struct rte_eth_fdir_info *fdir_info)
{
struct rte_eth_fdir_masks *mask =
&dev->data->dev_conf.fdir_conf.mask;
* 0 on success, a negative errno value otherwise and rte_errno is set.
*/
static int
-mlx5_fdir_ctrl_func(struct rte_eth_dev *dev, enum rte_filter_op filter_op,
+flow_fdir_ctrl_func(struct rte_eth_dev *dev, enum rte_filter_op filter_op,
void *arg)
{
enum rte_fdir_mode fdir_mode =
}
switch (filter_op) {
case RTE_ETH_FILTER_ADD:
- return mlx5_fdir_filter_add(dev, arg);
+ return flow_fdir_filter_add(dev, arg);
case RTE_ETH_FILTER_UPDATE:
- return mlx5_fdir_filter_update(dev, arg);
+ return flow_fdir_filter_update(dev, arg);
case RTE_ETH_FILTER_DELETE:
- return mlx5_fdir_filter_delete(dev, arg);
+ return flow_fdir_filter_delete(dev, arg);
case RTE_ETH_FILTER_FLUSH:
- mlx5_fdir_filter_flush(dev);
+ flow_fdir_filter_flush(dev);
break;
case RTE_ETH_FILTER_INFO:
- mlx5_fdir_info_get(dev, arg);
+ flow_fdir_info_get(dev, arg);
break;
default:
DRV_LOG(DEBUG, "port %u unknown operation %u",
*(const void **)arg = &mlx5_flow_ops;
return 0;
case RTE_ETH_FILTER_FDIR:
- return mlx5_fdir_ctrl_func(dev, filter_op, arg);
+ return flow_fdir_ctrl_func(dev, filter_op, arg);
default:
DRV_LOG(ERR, "port %u filter type (%d) not supported",
dev->data->port_id, filter_type);
}
return 0;
}
+
+#define MLX5_POOL_QUERY_FREQ_US 1000000
+
+/**
+ * Set the periodic procedure for triggering asynchronous batch queries for all
+ * the counter pools.
+ *
+ * @param[in] sh
+ * Pointer to mlx5_ibv_shared object.
+ */
+void
+mlx5_set_query_alarm(struct mlx5_ibv_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;
+
+ cont = MLX5_CNT_CONTAINER(sh, 1, 0);
+ pools_n += rte_atomic16_read(&cont->n_valid);
+ us = MLX5_POOL_QUERY_FREQ_US / pools_n;
+ DRV_LOG(DEBUG, "Set alarm for %u pools each %u us\n", pools_n, us);
+ if (rte_eal_alarm_set(us, mlx5_flow_query_alarm, sh)) {
+ sh->cmng.query_thread_on = 0;
+ DRV_LOG(ERR, "Cannot reinitialize query alarm\n");
+ } else {
+ sh->cmng.query_thread_on = 1;
+ }
+}
+
+/**
+ * The periodic procedure for triggering asynchronous batch queries for all the
+ * counter pools. This function is probably called by the host thread.
+ *
+ * @param[in] arg
+ * The parameter for the alarm process.
+ */
+void
+mlx5_flow_query_alarm(void *arg)
+{
+ struct mlx5_ibv_shared *sh = arg;
+ struct mlx5_devx_obj *dcs;
+ uint16_t offset;
+ int ret;
+ uint8_t batch = sh->cmng.batch;
+ uint16_t pool_index = sh->cmng.pool_index;
+ struct mlx5_pools_container *cont;
+ struct mlx5_pools_container *mcont;
+ struct mlx5_flow_counter_pool *pool;
+
+ 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;
+ }
+ if (!cont->pools) {
+ /* 2 empty containers case is unexpected. */
+ if (unlikely(batch != sh->cmng.batch))
+ goto set_alarm;
+ batch ^= 0x1;
+ pool_index = 0;
+ goto next_container;
+ }
+ pool = cont->pools[pool_index];
+ if (pool->raw_hw)
+ /* There is a pool query in progress. */
+ goto set_alarm;
+ pool->raw_hw =
+ LIST_FIRST(&sh->cmng.free_stat_raws);
+ if (!pool->raw_hw)
+ /* No free counter statistics raw memory. */
+ goto set_alarm;
+ dcs = (struct mlx5_devx_obj *)(uintptr_t)rte_atomic64_read
+ (&pool->a64_dcs);
+ offset = batch ? 0 : dcs->id % MLX5_COUNTERS_PER_POOL;
+ ret = mlx5_devx_cmd_flow_counter_query(dcs, 0, MLX5_COUNTERS_PER_POOL -
+ offset, NULL, NULL,
+ pool->raw_hw->mem_mng->dm->id,
+ (void *)(uintptr_t)
+ (pool->raw_hw->data + offset),
+ sh->devx_comp,
+ (uint64_t)(uintptr_t)pool);
+ if (ret) {
+ DRV_LOG(ERR, "Failed to trigger asynchronous query for dcs ID"
+ " %d\n", pool->min_dcs->id);
+ pool->raw_hw = NULL;
+ goto set_alarm;
+ }
+ pool->raw_hw->min_dcs_id = dcs->id;
+ LIST_REMOVE(pool->raw_hw, next);
+ sh->cmng.pending_queries++;
+ pool_index++;
+ if (pool_index >= rte_atomic16_read(&cont->n_valid)) {
+ batch ^= 0x1;
+ pool_index = 0;
+ }
+set_alarm:
+ sh->cmng.batch = batch;
+ sh->cmng.pool_index = pool_index;
+ mlx5_set_query_alarm(sh);
+}
+
+/**
+ * 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.
+ * @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,
+ 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;
+
+ if (unlikely(status)) {
+ raw_to_free = pool->raw_hw;
+ } else {
+ raw_to_free = pool->raw;
+ rte_spinlock_lock(&pool->sl);
+ pool->raw = pool->raw_hw;
+ rte_spinlock_unlock(&pool->sl);
+ rte_atomic64_add(&pool->query_gen, 1);
+ /* Be sure the new raw counters data is updated in memory. */
+ rte_cio_wmb();
+ }
+ LIST_INSERT_HEAD(&sh->cmng.free_stat_raws, raw_to_free, next);
+ pool->raw_hw = NULL;
+ sh->cmng.pending_queries--;
+}
+
+/**
+ * Translate the rte_flow group index to HW table value.
+ *
+ * @param[in] attributes
+ * Pointer to flow attributes
+ * @param[in] external
+ * Value is part of flow rule created by request external to PMD.
+ * @param[in] group
+ * rte_flow group index value.
+ * @param[out] table
+ * HW table value.
+ * @param[out] error
+ * Pointer to error structure.
+ *
+ * @return
+ * 0 on success, a negative errno value otherwise and rte_errno is set.
+ */
+int
+mlx5_flow_group_to_table(const struct rte_flow_attr *attributes, bool external,
+ uint32_t group, uint32_t *table,
+ struct rte_flow_error *error)
+{
+ if (attributes->transfer && external) {
+ if (group == UINT32_MAX)
+ return rte_flow_error_set
+ (error, EINVAL,
+ RTE_FLOW_ERROR_TYPE_ATTR_GROUP,
+ NULL,
+ "group index not supported");
+ *table = group + 1;
+ } else {
+ *table = group;
+ }
+ return 0;
+}