[PATCH 3/4] net/mlx5: add cross port CT object sharing

[Dariusz Sosnowski <dsosnowski@nvidia.com>]

From: Suanming Mou <suanmingm@nvidia.com>

This commit adds cross port CT object sharing.

Shared CT object shares the same DevX objects, but allocate port's
own action locally. Once the CT object is shared between two flows
in different ports, the two flows use their own local action with
the same offset index.

The shared CT object can only be created/updated/queried/destroyed
by host port.

Signed-off-by: Suanming Mou <suanmingm@nvidia.com>
Signed-off-by: Dariusz Sosnowski <dsosnowski@nvidia.com>
Acked-by: Ori Kam <orika@nvidia.com>
---
 doc/guides/rel_notes/release_24_03.rst |   3 +
 drivers/net/mlx5/mlx5_flow_hw.c        | 145 ++++++++++++++-----------
 2 files changed, 86 insertions(+), 62 deletions(-)

diff --git a/doc/guides/rel_notes/release_24_03.rst b/doc/guides/rel_notes/release_24_03.rst
index 619459baae..2262f350b5 100644
--- a/doc/guides/rel_notes/release_24_03.rst
+++ b/doc/guides/rel_notes/release_24_03.rst
@@ -133,6 +133,9 @@ New Features
   * Added HW steering support for modify field ``RTE_FLOW_FIELD_ESP_SEQ_NUM`` flow action.
   * Added HW steering support for modify field ``RTE_FLOW_FIELD_ESP_PROTO`` flow action.

+  * Added support for sharing indirect action objects of type ``RTE_FLOW_ACTION_TYPE_CONNTRACK``
+    with HW steering flow engine.
+

 Removed Items
 -------------
diff --git a/drivers/net/mlx5/mlx5_flow_hw.c b/drivers/net/mlx5/mlx5_flow_hw.c
index 2f366e9078..89066e7214 100644
--- a/drivers/net/mlx5/mlx5_flow_hw.c
+++ b/drivers/net/mlx5/mlx5_flow_hw.c
@@ -564,7 +564,7 @@ flow_hw_ct_compile(struct rte_eth_dev *dev,
 	struct mlx5_aso_ct_action *ct;

 	ct = mlx5_ipool_get(priv->hws_ctpool->cts, MLX5_ACTION_CTX_CT_GET_IDX(idx));
-	if (!ct || mlx5_aso_ct_available(priv->sh, queue, ct))
+	if (!ct || (!priv->shared_host && mlx5_aso_ct_available(priv->sh, queue, ct)))
 		return -1;
 	rule_act->action = priv->hws_ctpool->dr_action;
 	rule_act->aso_ct.offset = ct->offset;
@@ -3835,9 +3835,11 @@ __flow_hw_pull_indir_action_comp(struct rte_eth_dev *dev,
 	if (ret_comp < n_res && priv->hws_mpool)
 		ret_comp += mlx5_aso_pull_completion(&priv->hws_mpool->sq[queue],
 				&res[ret_comp], n_res - ret_comp);
-	if (ret_comp < n_res && priv->hws_ctpool)
-		ret_comp += mlx5_aso_pull_completion(&priv->ct_mng->aso_sqs[queue],
-				&res[ret_comp], n_res - ret_comp);
+	if (!priv->shared_host) {
+		if (ret_comp < n_res && priv->hws_ctpool)
+			ret_comp += mlx5_aso_pull_completion(&priv->ct_mng->aso_sqs[queue],
+					&res[ret_comp], n_res - ret_comp);
+	}
 	if (ret_comp < n_res && priv->quota_ctx.sq)
 		ret_comp += mlx5_aso_pull_completion(&priv->quota_ctx.sq[queue],
 						     &res[ret_comp],
@@ -8797,15 +8799,19 @@ flow_hw_ct_mng_destroy(struct rte_eth_dev *dev,
 }

 static void
-flow_hw_ct_pool_destroy(struct rte_eth_dev *dev __rte_unused,
+flow_hw_ct_pool_destroy(struct rte_eth_dev *dev,
 			struct mlx5_aso_ct_pool *pool)
 {
+	struct mlx5_priv *priv = dev->data->dev_private;
+
 	if (pool->dr_action)
 		mlx5dr_action_destroy(pool->dr_action);
-	if (pool->devx_obj)
-		claim_zero(mlx5_devx_cmd_destroy(pool->devx_obj));
-	if (pool->cts)
-		mlx5_ipool_destroy(pool->cts);
+	if (!priv->shared_host) {
+		if (pool->devx_obj)
+			claim_zero(mlx5_devx_cmd_destroy(pool->devx_obj));
+		if (pool->cts)
+			mlx5_ipool_destroy(pool->cts);
+	}
 	mlx5_free(pool);
 }

@@ -8829,51 +8835,56 @@ flow_hw_ct_pool_create(struct rte_eth_dev *dev,
 		.type = "mlx5_hw_ct_action",
 	};
 	int reg_id;
-	uint32_t flags;
+	uint32_t flags = 0;

-	if (port_attr->flags & RTE_FLOW_PORT_FLAG_SHARE_INDIRECT) {
-		DRV_LOG(ERR, "Connection tracking is not supported "
-			     "in cross vHCA sharing mode");
-		rte_errno = ENOTSUP;
-		return NULL;
-	}
 	pool = mlx5_malloc(MLX5_MEM_ZERO, sizeof(*pool), 0, SOCKET_ID_ANY);
 	if (!pool) {
 		rte_errno = ENOMEM;
 		return NULL;
 	}
-	obj = mlx5_devx_cmd_create_conn_track_offload_obj(priv->sh->cdev->ctx,
-							  priv->sh->cdev->pdn,
-							  log_obj_size);
-	if (!obj) {
-		rte_errno = ENODATA;
-		DRV_LOG(ERR, "Failed to create conn_track_offload_obj using DevX.");
-		goto err;
+	if (!priv->shared_host) {
+		/*
+		 * No need for local cache if CT number is a small number. Since
+		 * flow insertion rate will be very limited in that case. Here let's
+		 * set the number to less than default trunk size 4K.
+		 */
+		if (nb_cts <= cfg.trunk_size) {
+			cfg.per_core_cache = 0;
+			cfg.trunk_size = nb_cts;
+		} else if (nb_cts <= MLX5_HW_IPOOL_SIZE_THRESHOLD) {
+			cfg.per_core_cache = MLX5_HW_IPOOL_CACHE_MIN;
+		}
+		cfg.max_idx = nb_cts;
+		pool->cts = mlx5_ipool_create(&cfg);
+		if (!pool->cts)
+			goto err;
+		obj = mlx5_devx_cmd_create_conn_track_offload_obj(priv->sh->cdev->ctx,
+								  priv->sh->cdev->pdn,
+								  log_obj_size);
+		if (!obj) {
+			rte_errno = ENODATA;
+			DRV_LOG(ERR, "Failed to create conn_track_offload_obj using DevX.");
+			goto err;
+		}
+		pool->devx_obj = obj;
+	} else {
+		struct rte_eth_dev *host_dev = priv->shared_host;
+		struct mlx5_priv *host_priv = host_dev->data->dev_private;
+
+		pool->devx_obj = host_priv->hws_ctpool->devx_obj;
+		pool->cts = host_priv->hws_ctpool->cts;
+		MLX5_ASSERT(pool->cts);
+		MLX5_ASSERT(!port_attr->nb_conn_tracks);
 	}
-	pool->devx_obj = obj;
 	reg_id = mlx5_flow_get_reg_id(dev, MLX5_ASO_CONNTRACK, 0, NULL);
-	flags = MLX5DR_ACTION_FLAG_HWS_RX | MLX5DR_ACTION_FLAG_HWS_TX;
+	flags |= MLX5DR_ACTION_FLAG_HWS_RX | MLX5DR_ACTION_FLAG_HWS_TX;
 	if (priv->sh->config.dv_esw_en && priv->master)
 		flags |= MLX5DR_ACTION_FLAG_HWS_FDB;
 	pool->dr_action = mlx5dr_action_create_aso_ct(priv->dr_ctx,
-						      (struct mlx5dr_devx_obj *)obj,
+						      (struct mlx5dr_devx_obj *)pool->devx_obj,
 						      reg_id - REG_C_0, flags);
 	if (!pool->dr_action)
 		goto err;
-	/*
-	 * No need for local cache if CT number is a small number. Since
-	 * flow insertion rate will be very limited in that case. Here let's
-	 * set the number to less than default trunk size 4K.
-	 */
-	if (nb_cts <= cfg.trunk_size) {
-		cfg.per_core_cache = 0;
-		cfg.trunk_size = nb_cts;
-	} else if (nb_cts <= MLX5_HW_IPOOL_SIZE_THRESHOLD) {
-		cfg.per_core_cache = MLX5_HW_IPOOL_CACHE_MIN;
-	}
-	pool->cts = mlx5_ipool_create(&cfg);
-	if (!pool->cts)
-		goto err;
 	pool->sq = priv->ct_mng->aso_sqs;
 	/* Assign the last extra ASO SQ as public SQ. */
 	pool->shared_sq = &priv->ct_mng->aso_sqs[priv->nb_queue - 1];
@@ -9686,14 +9697,16 @@ flow_hw_configure(struct rte_eth_dev *dev,
 	if (!priv->shared_host)
 		flow_hw_create_send_to_kernel_actions(priv);
 	if (port_attr->nb_conn_tracks || (host_priv && host_priv->hws_ctpool)) {
-		mem_size = sizeof(struct mlx5_aso_sq) * nb_q_updated +
-			   sizeof(*priv->ct_mng);
-		priv->ct_mng = mlx5_malloc(MLX5_MEM_ZERO, mem_size,
-					   RTE_CACHE_LINE_SIZE, SOCKET_ID_ANY);
-		if (!priv->ct_mng)
-			goto err;
-		if (mlx5_aso_ct_queue_init(priv->sh, priv->ct_mng, nb_q_updated))
-			goto err;
+		if (!priv->shared_host) {
+			mem_size = sizeof(struct mlx5_aso_sq) * nb_q_updated +
+				sizeof(*priv->ct_mng);
+			priv->ct_mng = mlx5_malloc(MLX5_MEM_ZERO, mem_size,
+						RTE_CACHE_LINE_SIZE, SOCKET_ID_ANY);
+			if (!priv->ct_mng)
+				goto err;
+			if (mlx5_aso_ct_queue_init(priv->sh, priv->ct_mng, nb_q_updated))
+				goto err;
+		}
 		priv->hws_ctpool = flow_hw_ct_pool_create(dev, port_attr);
 		if (!priv->hws_ctpool)
 			goto err;
@@ -9914,17 +9927,20 @@ flow_hw_clear_port_info(struct rte_eth_dev *dev)
 }

 static int
-flow_hw_conntrack_destroy(struct rte_eth_dev *dev __rte_unused,
+flow_hw_conntrack_destroy(struct rte_eth_dev *dev,
 			  uint32_t idx,
 			  struct rte_flow_error *error)
 {
-	uint16_t owner = (uint16_t)MLX5_ACTION_CTX_CT_GET_OWNER(idx);
 	uint32_t ct_idx = MLX5_ACTION_CTX_CT_GET_IDX(idx);
-	struct rte_eth_dev *owndev = &rte_eth_devices[owner];
-	struct mlx5_priv *priv = owndev->data->dev_private;
+	struct mlx5_priv *priv = dev->data->dev_private;
 	struct mlx5_aso_ct_pool *pool = priv->hws_ctpool;
 	struct mlx5_aso_ct_action *ct;

+	if (priv->shared_host)
+		return rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
+				NULL,
+				"CT destruction is not allowed to guest port");
 	ct = mlx5_ipool_get(pool->cts, ct_idx);
 	if (!ct) {
 		return rte_flow_error_set(error, EINVAL,
@@ -9947,14 +9963,13 @@ flow_hw_conntrack_query(struct rte_eth_dev *dev, uint32_t queue, uint32_t idx,
 	struct mlx5_priv *priv = dev->data->dev_private;
 	struct mlx5_aso_ct_pool *pool = priv->hws_ctpool;
 	struct mlx5_aso_ct_action *ct;
-	uint16_t owner = (uint16_t)MLX5_ACTION_CTX_CT_GET_OWNER(idx);
 	uint32_t ct_idx;

-	if (owner != PORT_ID(priv))
-		return rte_flow_error_set(error, EACCES,
+	if (priv->shared_host)
+		return rte_flow_error_set(error, ENOTSUP,
 				RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
 				NULL,
-				"Can't query CT object owned by another port");
+				"CT query is not allowed to guest port");
 	ct_idx = MLX5_ACTION_CTX_CT_GET_IDX(idx);
 	ct = mlx5_ipool_get(pool->cts, ct_idx);
 	if (!ct) {
@@ -9984,15 +9999,14 @@ flow_hw_conntrack_update(struct rte_eth_dev *dev, uint32_t queue,
 	struct mlx5_aso_ct_pool *pool = priv->hws_ctpool;
 	struct mlx5_aso_ct_action *ct;
 	const struct rte_flow_action_conntrack *new_prf;
-	uint16_t owner = (uint16_t)MLX5_ACTION_CTX_CT_GET_OWNER(idx);
 	uint32_t ct_idx;
 	int ret = 0;

-	if (PORT_ID(priv) != owner)
-		return rte_flow_error_set(error, EACCES,
-					  RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
-					  NULL,
-					  "Can't update CT object owned by another port");
+	if (priv->shared_host)
+		return rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
+				NULL,
+				"CT update is not allowed to guest port");
 	ct_idx = MLX5_ACTION_CTX_CT_GET_IDX(idx);
 	ct = mlx5_ipool_get(pool->cts, ct_idx);
 	if (!ct) {
@@ -10042,6 +10056,13 @@ flow_hw_conntrack_create(struct rte_eth_dev *dev, uint32_t queue,
 	int ret;
 	bool async = !!(queue != MLX5_HW_INV_QUEUE);

+	if (priv->shared_host) {
+		rte_flow_error_set(error, ENOTSUP,
+				RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
+				NULL,
+				"CT create is not allowed to guest port");
+		return NULL;
+	}
 	if (!pool) {
 		rte_flow_error_set(error, EINVAL,
 				   RTE_FLOW_ERROR_TYPE_ACTION, NULL,
--
2.25.1