DPDK patches and discussions
 help / color / mirror / Atom feed
From: Dmitry Kozlyuk <dkozlyuk@oss.nvidia.com>
To: <dev@dpdk.org>
Cc: Ori Kam <orika@oss.nvidia.com>, <stable@dpdk.org>,
	Matan Azrad <matan@oss.nvidia.com>,
	Viacheslav Ovsiienko <viacheslavo@oss.nvidia.com>
Subject: [dpdk-dev] [PATCH v4 4/6] net/mlx5: discover max flow priority using DevX
Date: Thu, 21 Oct 2021 09:35:01 +0300	[thread overview]
Message-ID: <20211021063503.3632732-5-dkozlyuk@nvidia.com> (raw)
In-Reply-To: <20211021063503.3632732-1-dkozlyuk@nvidia.com>

Maximum available flow priority was discovered using Verbs API
regardless of the selected flow engine. This required some Verbs
objects to be initialized in order to use DevX engine. Make priority
discovery an engine method and implement it for DevX using its API.

Cc: stable@dpdk.org

Signed-off-by: Dmitry Kozlyuk <dkozlyuk@nvidia.com>
Acked-by: Matan Azrad <matan@nvidia.com>
---
 drivers/net/mlx5/linux/mlx5_os.c   |   1 -
 drivers/net/mlx5/mlx5_flow.c       |  98 +++++++++++++++++++++++++++
 drivers/net/mlx5/mlx5_flow.h       |   4 ++
 drivers/net/mlx5/mlx5_flow_dv.c    | 103 +++++++++++++++++++++++++++++
 drivers/net/mlx5/mlx5_flow_verbs.c |  77 +++------------------
 5 files changed, 215 insertions(+), 68 deletions(-)

diff --git a/drivers/net/mlx5/linux/mlx5_os.c b/drivers/net/mlx5/linux/mlx5_os.c
index 26a8d75b99..60d2c398db 100644
--- a/drivers/net/mlx5/linux/mlx5_os.c
+++ b/drivers/net/mlx5/linux/mlx5_os.c
@@ -1849,7 +1849,6 @@ mlx5_dev_spawn(struct rte_device *dpdk_dev,
 	priv->drop_queue.hrxq = mlx5_drop_action_create(eth_dev);
 	if (!priv->drop_queue.hrxq)
 		goto error;
-	/* Supported Verbs flow priority number detection. */
 	err = mlx5_flow_discover_priorities(eth_dev);
 	if (err < 0) {
 		err = -err;
diff --git a/drivers/net/mlx5/mlx5_flow.c b/drivers/net/mlx5/mlx5_flow.c
index b4d0b7b5ef..2768244c2e 100644
--- a/drivers/net/mlx5/mlx5_flow.c
+++ b/drivers/net/mlx5/mlx5_flow.c
@@ -9480,3 +9480,101 @@ mlx5_flow_expand_rss_adjust_node(const struct rte_flow_item *pattern,
 		return node;
 	}
 }
+
+/* Map of Verbs to Flow priority with 8 Verbs priorities. */
+static const uint32_t priority_map_3[][MLX5_PRIORITY_MAP_MAX] = {
+	{ 0, 1, 2 }, { 2, 3, 4 }, { 5, 6, 7 },
+};
+
+/* Map of Verbs to Flow priority with 16 Verbs priorities. */
+static const uint32_t priority_map_5[][MLX5_PRIORITY_MAP_MAX] = {
+	{ 0, 1, 2 }, { 3, 4, 5 }, { 6, 7, 8 },
+	{ 9, 10, 11 }, { 12, 13, 14 },
+};
+
+/**
+ * Discover the number of available flow priorities.
+ *
+ * @param dev
+ *   Ethernet device.
+ *
+ * @return
+ *   On success, number of available flow priorities.
+ *   On failure, a negative errno-style code and rte_errno is set.
+ */
+int
+mlx5_flow_discover_priorities(struct rte_eth_dev *dev)
+{
+	static const uint16_t vprio[] = {8, 16};
+	const struct mlx5_priv *priv = dev->data->dev_private;
+	const struct mlx5_flow_driver_ops *fops;
+	enum mlx5_flow_drv_type type;
+	int ret;
+
+	type = mlx5_flow_os_get_type();
+	if (type == MLX5_FLOW_TYPE_MAX) {
+		type = MLX5_FLOW_TYPE_VERBS;
+		if (priv->config.devx && priv->config.dv_flow_en)
+			type = MLX5_FLOW_TYPE_DV;
+	}
+	fops = flow_get_drv_ops(type);
+	if (fops->discover_priorities == NULL) {
+		DRV_LOG(ERR, "Priority discovery not supported");
+		rte_errno = ENOTSUP;
+		return -rte_errno;
+	}
+	ret = fops->discover_priorities(dev, vprio, RTE_DIM(vprio));
+	if (ret < 0)
+		return ret;
+	switch (ret) {
+	case 8:
+		ret = RTE_DIM(priority_map_3);
+		break;
+	case 16:
+		ret = RTE_DIM(priority_map_5);
+		break;
+	default:
+		rte_errno = ENOTSUP;
+		DRV_LOG(ERR,
+			"port %u maximum priority: %d expected 8/16",
+			dev->data->port_id, ret);
+		return -rte_errno;
+	}
+	DRV_LOG(INFO, "port %u supported flow priorities:"
+		" 0-%d for ingress or egress root table,"
+		" 0-%d for non-root table or transfer root table.",
+		dev->data->port_id, ret - 2,
+		MLX5_NON_ROOT_FLOW_MAX_PRIO - 1);
+	return ret;
+}
+
+/**
+ * Adjust flow priority based on the highest layer and the request priority.
+ *
+ * @param[in] dev
+ *   Pointer to the Ethernet device structure.
+ * @param[in] priority
+ *   The rule base priority.
+ * @param[in] subpriority
+ *   The priority based on the items.
+ *
+ * @return
+ *   The new priority.
+ */
+uint32_t
+mlx5_flow_adjust_priority(struct rte_eth_dev *dev, int32_t priority,
+			  uint32_t subpriority)
+{
+	uint32_t res = 0;
+	struct mlx5_priv *priv = dev->data->dev_private;
+
+	switch (priv->config.flow_prio) {
+	case RTE_DIM(priority_map_3):
+		res = priority_map_3[priority][subpriority];
+		break;
+	case RTE_DIM(priority_map_5):
+		res = priority_map_5[priority][subpriority];
+		break;
+	}
+	return  res;
+}
diff --git a/drivers/net/mlx5/mlx5_flow.h b/drivers/net/mlx5/mlx5_flow.h
index 5c68d4f7d7..8f94125f26 100644
--- a/drivers/net/mlx5/mlx5_flow.h
+++ b/drivers/net/mlx5/mlx5_flow.h
@@ -1226,6 +1226,9 @@ typedef int (*mlx5_flow_create_def_policy_t)
 			(struct rte_eth_dev *dev);
 typedef void (*mlx5_flow_destroy_def_policy_t)
 			(struct rte_eth_dev *dev);
+typedef int (*mlx5_flow_discover_priorities_t)
+			(struct rte_eth_dev *dev,
+			 const uint16_t *vprio, int vprio_n);
 
 struct mlx5_flow_driver_ops {
 	mlx5_flow_validate_t validate;
@@ -1260,6 +1263,7 @@ struct mlx5_flow_driver_ops {
 	mlx5_flow_action_update_t action_update;
 	mlx5_flow_action_query_t action_query;
 	mlx5_flow_sync_domain_t sync_domain;
+	mlx5_flow_discover_priorities_t discover_priorities;
 };
 
 /* mlx5_flow.c */
diff --git a/drivers/net/mlx5/mlx5_flow_dv.c b/drivers/net/mlx5/mlx5_flow_dv.c
index e31d4d8468..5163f518d7 100644
--- a/drivers/net/mlx5/mlx5_flow_dv.c
+++ b/drivers/net/mlx5/mlx5_flow_dv.c
@@ -17883,6 +17883,108 @@ flow_dv_sync_domain(struct rte_eth_dev *dev, uint32_t domains, uint32_t flags)
 	return 0;
 }
 
+/**
+ * Discover the number of available flow priorities
+ * by trying to create a flow with the highest priority value
+ * for each possible number.
+ *
+ * @param[in] dev
+ *   Ethernet device.
+ * @param[in] vprio
+ *   List of possible number of available priorities.
+ * @param[in] vprio_n
+ *   Size of @p vprio array.
+ * @return
+ *   On success, number of available flow priorities.
+ *   On failure, a negative errno-style code and rte_errno is set.
+ */
+static int
+flow_dv_discover_priorities(struct rte_eth_dev *dev,
+			    const uint16_t *vprio, int vprio_n)
+{
+	struct mlx5_priv *priv = dev->data->dev_private;
+	struct mlx5_indexed_pool *pool = priv->sh->ipool[MLX5_IPOOL_MLX5_FLOW];
+	struct rte_flow_item_eth eth;
+	struct rte_flow_item item = {
+		.type = RTE_FLOW_ITEM_TYPE_ETH,
+		.spec = &eth,
+		.mask = &eth,
+	};
+	struct mlx5_flow_dv_matcher matcher = {
+		.mask = {
+			.size = sizeof(matcher.mask.buf),
+		},
+	};
+	union mlx5_flow_tbl_key tbl_key;
+	struct mlx5_flow flow;
+	void *action;
+	struct rte_flow_error error;
+	uint8_t misc_mask;
+	int i, err, ret = -ENOTSUP;
+
+	/*
+	 * Prepare a flow with a catch-all pattern and a drop action.
+	 * Use drop queue, because shared drop action may be unavailable.
+	 */
+	action = priv->drop_queue.hrxq->action;
+	if (action == NULL) {
+		DRV_LOG(ERR, "Priority discovery requires a drop action");
+		rte_errno = ENOTSUP;
+		return -rte_errno;
+	}
+	memset(&flow, 0, sizeof(flow));
+	flow.handle = mlx5_ipool_zmalloc(pool, &flow.handle_idx);
+	if (flow.handle == NULL) {
+		DRV_LOG(ERR, "Cannot create flow handle");
+		rte_errno = ENOMEM;
+		return -rte_errno;
+	}
+	flow.ingress = true;
+	flow.dv.value.size = MLX5_ST_SZ_BYTES(fte_match_param);
+	flow.dv.actions[0] = action;
+	flow.dv.actions_n = 1;
+	memset(&eth, 0, sizeof(eth));
+	flow_dv_translate_item_eth(matcher.mask.buf, flow.dv.value.buf,
+				   &item, /* inner */ false, /* group */ 0);
+	matcher.crc = rte_raw_cksum(matcher.mask.buf, matcher.mask.size);
+	for (i = 0; i < vprio_n; i++) {
+		/* Configure the next proposed maximum priority. */
+		matcher.priority = vprio[i] - 1;
+		memset(&tbl_key, 0, sizeof(tbl_key));
+		err = flow_dv_matcher_register(dev, &matcher, &tbl_key, &flow,
+					       /* tunnel */ NULL,
+					       /* group */ 0,
+					       &error);
+		if (err != 0) {
+			/* This action is pure SW and must always succeed. */
+			DRV_LOG(ERR, "Cannot register matcher");
+			ret = -rte_errno;
+			break;
+		}
+		/* Try to apply the flow to HW. */
+		misc_mask = flow_dv_matcher_enable(flow.dv.value.buf);
+		__flow_dv_adjust_buf_size(&flow.dv.value.size, misc_mask);
+		err = mlx5_flow_os_create_flow
+				(flow.handle->dvh.matcher->matcher_object,
+				 (void *)&flow.dv.value, flow.dv.actions_n,
+				 flow.dv.actions, &flow.handle->drv_flow);
+		if (err == 0) {
+			claim_zero(mlx5_flow_os_destroy_flow
+						(flow.handle->drv_flow));
+			flow.handle->drv_flow = NULL;
+		}
+		claim_zero(flow_dv_matcher_release(dev, flow.handle));
+		if (err != 0)
+			break;
+		ret = vprio[i];
+	}
+	mlx5_ipool_free(pool, flow.handle_idx);
+	/* Set rte_errno if no expected priority value matched. */
+	if (ret < 0)
+		rte_errno = -ret;
+	return ret;
+}
+
 const struct mlx5_flow_driver_ops mlx5_flow_dv_drv_ops = {
 	.validate = flow_dv_validate,
 	.prepare = flow_dv_prepare,
@@ -17916,6 +18018,7 @@ const struct mlx5_flow_driver_ops mlx5_flow_dv_drv_ops = {
 	.action_update = flow_dv_action_update,
 	.action_query = flow_dv_action_query,
 	.sync_domain = flow_dv_sync_domain,
+	.discover_priorities = flow_dv_discover_priorities,
 };
 
 #endif /* HAVE_IBV_FLOW_DV_SUPPORT */
diff --git a/drivers/net/mlx5/mlx5_flow_verbs.c b/drivers/net/mlx5/mlx5_flow_verbs.c
index 1627c3905f..c8cf7ef29c 100644
--- a/drivers/net/mlx5/mlx5_flow_verbs.c
+++ b/drivers/net/mlx5/mlx5_flow_verbs.c
@@ -28,17 +28,6 @@
 #define VERBS_SPEC_INNER(item_flags) \
 	(!!((item_flags) & MLX5_FLOW_LAYER_TUNNEL) ? IBV_FLOW_SPEC_INNER : 0)
 
-/* Map of Verbs to Flow priority with 8 Verbs priorities. */
-static const uint32_t priority_map_3[][MLX5_PRIORITY_MAP_MAX] = {
-	{ 0, 1, 2 }, { 2, 3, 4 }, { 5, 6, 7 },
-};
-
-/* Map of Verbs to Flow priority with 16 Verbs priorities. */
-static const uint32_t priority_map_5[][MLX5_PRIORITY_MAP_MAX] = {
-	{ 0, 1, 2 }, { 3, 4, 5 }, { 6, 7, 8 },
-	{ 9, 10, 11 }, { 12, 13, 14 },
-};
-
 /* Verbs specification header. */
 struct ibv_spec_header {
 	enum ibv_flow_spec_type type;
@@ -50,13 +39,17 @@ struct ibv_spec_header {
  *
  * @param[in] dev
  *   Pointer to the Ethernet device structure.
- *
+ * @param[in] vprio
+ *   Expected result variants.
+ * @param[in] vprio_n
+ *   Number of entries in @p vprio array.
  * @return
- *   number of supported flow priority on success, a negative errno
+ *   Number of supported flow priority on success, a negative errno
  *   value otherwise and rte_errno is set.
  */
-int
-mlx5_flow_discover_priorities(struct rte_eth_dev *dev)
+static int
+flow_verbs_discover_priorities(struct rte_eth_dev *dev,
+			       const uint16_t *vprio, int vprio_n)
 {
 	struct mlx5_priv *priv = dev->data->dev_private;
 	struct {
@@ -79,7 +72,6 @@ mlx5_flow_discover_priorities(struct rte_eth_dev *dev)
 	};
 	struct ibv_flow *flow;
 	struct mlx5_hrxq *drop = priv->drop_queue.hrxq;
-	uint16_t vprio[] = { 8, 16 };
 	int i;
 	int priority = 0;
 
@@ -87,7 +79,7 @@ mlx5_flow_discover_priorities(struct rte_eth_dev *dev)
 		rte_errno = ENOTSUP;
 		return -rte_errno;
 	}
-	for (i = 0; i != RTE_DIM(vprio); i++) {
+	for (i = 0; i != vprio_n; i++) {
 		flow_attr.attr.priority = vprio[i] - 1;
 		flow = mlx5_glue->create_flow(drop->qp, &flow_attr.attr);
 		if (!flow)
@@ -95,59 +87,9 @@ mlx5_flow_discover_priorities(struct rte_eth_dev *dev)
 		claim_zero(mlx5_glue->destroy_flow(flow));
 		priority = vprio[i];
 	}
-	switch (priority) {
-	case 8:
-		priority = RTE_DIM(priority_map_3);
-		break;
-	case 16:
-		priority = RTE_DIM(priority_map_5);
-		break;
-	default:
-		rte_errno = ENOTSUP;
-		DRV_LOG(ERR,
-			"port %u verbs maximum priority: %d expected 8/16",
-			dev->data->port_id, priority);
-		return -rte_errno;
-	}
-	DRV_LOG(INFO, "port %u supported flow priorities:"
-		" 0-%d for ingress or egress root table,"
-		" 0-%d for non-root table or transfer root table.",
-		dev->data->port_id, priority - 2,
-		MLX5_NON_ROOT_FLOW_MAX_PRIO - 1);
 	return priority;
 }
 
-/**
- * Adjust flow priority based on the highest layer and the request priority.
- *
- * @param[in] dev
- *   Pointer to the Ethernet device structure.
- * @param[in] priority
- *   The rule base priority.
- * @param[in] subpriority
- *   The priority based on the items.
- *
- * @return
- *   The new priority.
- */
-uint32_t
-mlx5_flow_adjust_priority(struct rte_eth_dev *dev, int32_t priority,
-				   uint32_t subpriority)
-{
-	uint32_t res = 0;
-	struct mlx5_priv *priv = dev->data->dev_private;
-
-	switch (priv->config.flow_prio) {
-	case RTE_DIM(priority_map_3):
-		res = priority_map_3[priority][subpriority];
-		break;
-	case RTE_DIM(priority_map_5):
-		res = priority_map_5[priority][subpriority];
-		break;
-	}
-	return  res;
-}
-
 /**
  * Get Verbs flow counter by index.
  *
@@ -2087,4 +2029,5 @@ const struct mlx5_flow_driver_ops mlx5_flow_verbs_drv_ops = {
 	.destroy = flow_verbs_destroy,
 	.query = flow_verbs_query,
 	.sync_domain = flow_verbs_sync_domain,
+	.discover_priorities = flow_verbs_discover_priorities,
 };
-- 
2.25.1


  parent reply	other threads:[~2021-10-21  6:36 UTC|newest]

Thread overview: 96+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-10-05  0:52 [dpdk-dev] [PATCH 0/5] Flow entites behavior on port restart dkozlyuk
2021-10-05  0:52 ` [dpdk-dev] [PATCH 1/5] ethdev: add capability to keep flow rules on restart dkozlyuk
2021-10-06  6:15   ` Ori Kam
2021-10-06  6:55     ` Somnath Kotur
2021-10-06 17:15   ` Ajit Khaparde
2021-10-05  0:52 ` [dpdk-dev] [PATCH 2/5] ethdev: add capability to keep shared objects " dkozlyuk
2021-10-06  6:16   ` Ori Kam
2021-10-13  8:32   ` Dmitry Kozlyuk
2021-10-14 13:46     ` Ferruh Yigit
2021-10-14 21:45       ` Dmitry Kozlyuk
2021-10-14 21:48         ` Dmitry Kozlyuk
2021-10-15 11:46         ` Ferruh Yigit
2021-10-15 12:35           ` Dmitry Kozlyuk
2021-10-15 16:26             ` Ferruh Yigit
2021-10-16 20:32               ` Dmitry Kozlyuk
2021-10-18  8:42                 ` Ferruh Yigit
2021-10-18 11:13                   ` Dmitry Kozlyuk
2021-10-18 11:59                     ` Ferruh Yigit
2021-10-14 14:14     ` Dmitry Kozlyuk
2021-10-15  8:26       ` Andrew Rybchenko
2021-10-15  9:04         ` Dmitry Kozlyuk
2021-10-15  9:36           ` Andrew Rybchenko
2021-10-05  0:52 ` [dpdk-dev] [PATCH 3/5] net/mlx5: discover max flow priority using DevX dkozlyuk
2021-10-05  0:52 ` [dpdk-dev] [PATCH 4/5] net/mlx5: create drop queue " dkozlyuk
2021-10-05  0:52 ` [dpdk-dev] [PATCH 5/5] net/mlx5: preserve indirect actions on restart dkozlyuk
2021-10-15 16:18 ` [dpdk-dev] [PATCH v2 0/5] Flow entites behavior on port restart Dmitry Kozlyuk
2021-10-15 16:18   ` [dpdk-dev] [PATCH v2 1/5] ethdev: add capability to keep flow rules on restart Dmitry Kozlyuk
2021-10-18  8:56     ` Andrew Rybchenko
2021-10-19 12:38       ` Dmitry Kozlyuk
2021-10-18 13:06     ` Zhang, Qi Z
2021-10-18 22:51       ` Dmitry Kozlyuk
2021-10-19  1:00         ` Zhang, Qi Z
2021-10-15 16:18   ` [dpdk-dev] [PATCH v2 2/5] ethdev: add capability to keep shared objects " Dmitry Kozlyuk
2021-10-17  8:10     ` Ori Kam
2021-10-17  9:14       ` Dmitry Kozlyuk
2021-10-17  9:45         ` Ori Kam
2021-10-15 16:18   ` [dpdk-dev] [PATCH v2 3/5] net/mlx5: discover max flow priority using DevX Dmitry Kozlyuk
2021-10-15 16:18   ` [dpdk-dev] [PATCH v2 4/5] net/mlx5: create drop queue " Dmitry Kozlyuk
2021-10-15 16:18   ` [dpdk-dev] [PATCH v2 5/5] net/mlx5: preserve indirect actions on restart Dmitry Kozlyuk
2021-10-19 12:37   ` [dpdk-dev] [PATCH v3 0/6] Flow entites behavior on port restart Dmitry Kozlyuk
2021-10-19 12:37     ` [dpdk-dev] [PATCH v3 1/6] ethdev: add capability to keep flow rules on restart Dmitry Kozlyuk
2021-10-19 15:22       ` Ori Kam
2021-10-19 16:38       ` Ferruh Yigit
2021-10-19 17:13         ` Dmitry Kozlyuk
2021-10-20 10:39       ` Andrew Rybchenko
2021-10-20 11:40         ` Dmitry Kozlyuk
2021-10-20 13:40           ` Ori Kam
2021-10-19 12:37     ` [dpdk-dev] [PATCH v3 2/6] ethdev: add capability to keep shared objects " Dmitry Kozlyuk
2021-10-19 15:22       ` Ori Kam
2021-10-19 12:37     ` [dpdk-dev] [PATCH v3 3/6] net: advertise no support for keeping flow rules Dmitry Kozlyuk
2021-10-20 10:08       ` Andrew Rybchenko
2021-10-20 22:20         ` Dmitry Kozlyuk
2021-10-19 12:37     ` [dpdk-dev] [PATCH v3 4/6] net/mlx5: discover max flow priority using DevX Dmitry Kozlyuk
2021-10-19 12:37     ` [dpdk-dev] [PATCH v3 5/6] net/mlx5: create drop queue " Dmitry Kozlyuk
2021-10-19 12:37     ` [dpdk-dev] [PATCH v3 6/6] net/mlx5: preserve indirect actions on restart Dmitry Kozlyuk
2021-10-20 10:12     ` [dpdk-dev] [PATCH v3 0/6] Flow entites behavior on port restart Andrew Rybchenko
2021-10-20 13:21       ` Dmitry Kozlyuk
2021-10-21  6:34     ` [dpdk-dev] [PATCH v4 " Dmitry Kozlyuk
2021-10-21  6:34       ` [dpdk-dev] [PATCH v4 1/6] ethdev: add capability to keep flow rules on restart Dmitry Kozlyuk
2021-10-21  7:36         ` Ori Kam
2021-10-28 18:33         ` Ajit Khaparde
2021-11-01 15:02         ` Andrew Rybchenko
2021-11-01 15:56           ` Dmitry Kozlyuk
2021-10-21  6:34       ` [dpdk-dev] [PATCH v4 2/6] ethdev: add capability to keep shared objects " Dmitry Kozlyuk
2021-10-21  7:37         ` Ori Kam
2021-10-21 18:28         ` Ajit Khaparde
2021-11-01 15:04         ` Andrew Rybchenko
2021-10-21  6:35       ` [dpdk-dev] [PATCH v4 3/6] net: advertise no support for keeping flow rules Dmitry Kozlyuk
2021-10-21 18:26         ` Ajit Khaparde
2021-10-22  1:38           ` Somnath Kotur
2021-10-27  7:11         ` Hyong Youb Kim (hyonkim)
2021-11-01 15:06         ` Andrew Rybchenko
2021-11-01 16:59           ` Ferruh Yigit
2021-10-21  6:35       ` Dmitry Kozlyuk [this message]
2021-10-21  6:35       ` [dpdk-dev] [PATCH v4 5/6] net/mlx5: create drop queue using DevX Dmitry Kozlyuk
2021-10-21  6:35       ` [dpdk-dev] [PATCH v4 6/6] net/mlx5: preserve indirect actions on restart Dmitry Kozlyuk
2021-10-26 11:46       ` [dpdk-dev] [PATCH v4 0/6] Flow entites behavior on port restart Ferruh Yigit
2021-11-01 13:43         ` Ferruh Yigit
2021-11-02 13:49       ` Ferruh Yigit
2021-11-02 13:54       ` [dpdk-dev] [PATCH v5 " Dmitry Kozlyuk
2021-11-02 13:54         ` [dpdk-dev] [PATCH v5 1/6] ethdev: add capability to keep flow rules on restart Dmitry Kozlyuk
2021-11-02 13:54         ` [dpdk-dev] [PATCH v5 2/6] ethdev: add capability to keep shared objects " Dmitry Kozlyuk
2021-11-02 13:54         ` [dpdk-dev] [PATCH v5 3/6] net: advertise no support for keeping flow rules Dmitry Kozlyuk
2021-11-02 13:54         ` [dpdk-dev] [PATCH v5 4/6] net/mlx5: discover max flow priority using DevX Dmitry Kozlyuk
2021-11-02 13:54         ` [dpdk-dev] [PATCH v5 5/6] net/mlx5: create drop queue " Dmitry Kozlyuk
2021-11-02 13:54         ` [dpdk-dev] [PATCH v5 6/6] net/mlx5: preserve indirect actions on restart Dmitry Kozlyuk
2021-11-02 14:23         ` [dpdk-dev] [PATCH v5 0/6] Flow entites behavior on port restart Ferruh Yigit
2021-11-02 17:02           ` Dmitry Kozlyuk
2021-11-02 17:01         ` [dpdk-dev] [PATCH v6 " Dmitry Kozlyuk
2021-11-02 17:01           ` [dpdk-dev] [PATCH v6 1/6] ethdev: add capability to keep flow rules on restart Dmitry Kozlyuk
2021-11-02 17:01           ` [dpdk-dev] [PATCH v6 2/6] ethdev: add capability to keep shared objects " Dmitry Kozlyuk
2021-11-02 17:01           ` [dpdk-dev] [PATCH v6 3/6] net: advertise no support for keeping flow rules Dmitry Kozlyuk
2021-11-02 17:01           ` [dpdk-dev] [PATCH v6 4/6] net/mlx5: discover max flow priority using DevX Dmitry Kozlyuk
2021-11-02 17:01           ` [dpdk-dev] [PATCH v6 5/6] net/mlx5: create drop queue " Dmitry Kozlyuk
2021-11-02 17:01           ` [dpdk-dev] [PATCH v6 6/6] net/mlx5: preserve indirect actions on restart Dmitry Kozlyuk
2021-11-02 18:02           ` [dpdk-dev] [PATCH v6 0/6] Flow entites behavior on port restart Ferruh Yigit

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20211021063503.3632732-5-dkozlyuk@nvidia.com \
    --to=dkozlyuk@oss.nvidia.com \
    --cc=dev@dpdk.org \
    --cc=matan@oss.nvidia.com \
    --cc=orika@oss.nvidia.com \
    --cc=stable@dpdk.org \
    --cc=viacheslavo@oss.nvidia.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).