Merge branch 'devlink-rate-objects'

Dmytro Linkin says:

====================
devlink: rate objects API

Resending without RFC.

Currently kernel provides a way to change tx rate of single VF in
switchdev mode via tc-police action. When lots of VFs are configured
management of theirs rates becomes non-trivial task and some grouping
mechanism is required. Implementing such grouping in tc-police will bring
flow related limitations and unwanted complications, like:
- tc-police is a policer and there is a user request for a traffic
  shaper, so shared tc-police action is not suitable;
- flows requires net device to be placed on, means "groups" wouldn't
  have net device instance itself. Taking into the account previous
  point was reviewed a sollution, when representor have a policer and
  the driver use a shaper if qdisc contains group of VFs - such approach
  ugly, compilated and misleading;
- TC is ingress only, while configuring "other" side of the wire looks
  more like a "real" picture where shaping is outside of the steering
  world, similar to "ip link" command;

According to that devlink is the most appropriate place.

This series introduces devlink API for managing tx rate of single devlink
port or of a group by invoking callbacks (see below) of corresponding
driver. Also devlink port or a group can be added to the parent group,
where driver responsible to handle rates of a group elements. To achieve
all of that new rate object is added. It can be one of the two types:
- leaf - represents a single devlink port; created/destroyed by the
  driver and bound to the devlink port. As example, some driver may
  create leaf rate object for every devlink port associated with VF.
  Since leaf have 1to1 mapping to it's devlink port, in user space it is
  referred as pci/<bus_addr>/<port_index>;
- node - represents a group of rate objects; created/deleted by request
  from the userspace; initially empty (no rate objects added). In
  userspace it is referred as pci/<bus_addr>/<node_name>, where node name
  can be any, except decimal number, to avoid collisions with leafs.

devlink_ops extended with following callbacks:
- rate_{leaf|node}_tx_{share|max}_set
- rate_node_{new|del}
- rate_{leaf|node}_parent_set

KAPI provides:
- creation/destruction of the leaf rate object associated with devlink
  port
- destruction of rate nodes to allow a vendor driver to free allocated
  resources on driver removal or due to the other reasons when nodes
  destruction required

UAPI provides:
- dumping all or single rate objects
- setting tx_{share|max} of rate object of any type
- creating/deleting node rate object
- setting/unsetting parent of any rate object

Added devlink rate object support for netdevsim driver

Issues/open questions:
- Does user need DEVLINK_CMD_RATE_DEL_ALL_CHILD command to clean all
  children of particular parent node? For example:
  $ devlink port function rate flush netdevsim/netdevsim10/group
- priv pointer passed to the callbacks is a source of bugs; in leaf case
  driver can embed rate object into internal structure and use
  container_of() on it; in node case it cannot be done since nodes are
  created from userspace

v1->v2:
- fixed kernel-doc for devlink_rate_leaf_{create|destroy}()
- s/func/function/ for all devlink port command occurences

v2->v3:
- devlink:
  - added devlink_rate_nodes_destroy() function
- netdevsim:
  - added call of devlink_rate_nodes_destroy() function
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

1 files changed, 659 insertions, 1 deletions

diff --git a/net/core/devlink.c b/net/core/devlink.c
index 69681f19388e..464f56408247 100644
--- a/net/core/devlink.c
+++ b/net/core/devlink.c
@@ -190,6 +190,80 @@ static struct devlink_port *devlink_port_get_from_info(struct devlink *devlink,
 	return devlink_port_get_from_attrs(devlink, info->attrs);
 }
 
+static inline bool
+devlink_rate_is_leaf(struct devlink_rate *devlink_rate)
+{
+	return devlink_rate->type == DEVLINK_RATE_TYPE_LEAF;
+}
+
+static inline bool
+devlink_rate_is_node(struct devlink_rate *devlink_rate)
+{
+	return devlink_rate->type == DEVLINK_RATE_TYPE_NODE;
+}
+
+static struct devlink_rate *
+devlink_rate_leaf_get_from_info(struct devlink *devlink, struct genl_info *info)
+{
+	struct devlink_rate *devlink_rate;
+	struct devlink_port *devlink_port;
+
+	devlink_port = devlink_port_get_from_attrs(devlink, info->attrs);
+	if (IS_ERR(devlink_port))
+		return ERR_CAST(devlink_port);
+	devlink_rate = devlink_port->devlink_rate;
+	return devlink_rate ?: ERR_PTR(-ENODEV);
+}
+
+static struct devlink_rate *
+devlink_rate_node_get_by_name(struct devlink *devlink, const char *node_name)
+{
+	static struct devlink_rate *devlink_rate;
+
+	list_for_each_entry(devlink_rate, &devlink->rate_list, list) {
+		if (devlink_rate_is_node(devlink_rate) &&
+		    !strcmp(node_name, devlink_rate->name))
+			return devlink_rate;
+	}
+	return ERR_PTR(-ENODEV);
+}
+
+static struct devlink_rate *
+devlink_rate_node_get_from_attrs(struct devlink *devlink, struct nlattr **attrs)
+{
+	const char *rate_node_name;
+	size_t len;
+
+	if (!attrs[DEVLINK_ATTR_RATE_NODE_NAME])
+		return ERR_PTR(-EINVAL);
+	rate_node_name = nla_data(attrs[DEVLINK_ATTR_RATE_NODE_NAME]);
+	len = strlen(rate_node_name);
+	/* Name cannot be empty or decimal number */
+	if (!len || strspn(rate_node_name, "0123456789") == len)
+		return ERR_PTR(-EINVAL);
+
+	return devlink_rate_node_get_by_name(devlink, rate_node_name);
+}
+
+static struct devlink_rate *
+devlink_rate_node_get_from_info(struct devlink *devlink, struct genl_info *info)
+{
+	return devlink_rate_node_get_from_attrs(devlink, info->attrs);
+}
+
+static struct devlink_rate *
+devlink_rate_get_from_info(struct devlink *devlink, struct genl_info *info)
+{
+	struct nlattr **attrs = info->attrs;
+
+	if (attrs[DEVLINK_ATTR_PORT_INDEX])
+		return devlink_rate_leaf_get_from_info(devlink, info);
+	else if (attrs[DEVLINK_ATTR_RATE_NODE_NAME])
+		return devlink_rate_node_get_from_info(devlink, info);
+	else
+		return ERR_PTR(-EINVAL);
+}
+
 struct devlink_sb {
 	struct list_head list;
 	unsigned int index;
@@ -408,12 +482,14 @@ devlink_region_snapshot_get_by_id(struct devlink_region *region, u32 id)
 
 #define DEVLINK_NL_FLAG_NEED_PORT		BIT(0)
 #define DEVLINK_NL_FLAG_NEED_DEVLINK_OR_PORT	BIT(1)
+#define DEVLINK_NL_FLAG_NEED_RATE		BIT(2)
+#define DEVLINK_NL_FLAG_NEED_RATE_NODE		BIT(3)
 
 /* The per devlink instance lock is taken by default in the pre-doit
  * operation, yet several commands do not require this. The global
  * devlink lock is taken and protects from disruption by user-calls.
  */
-#define DEVLINK_NL_FLAG_NO_LOCK			BIT(2)
+#define DEVLINK_NL_FLAG_NO_LOCK			BIT(4)
 
 static int devlink_nl_pre_doit(const struct genl_ops *ops,
 			       struct sk_buff *skb, struct genl_info *info)
@@ -442,6 +518,24 @@ static int devlink_nl_pre_doit(const struct genl_ops *ops,
 		devlink_port = devlink_port_get_from_info(devlink, info);
 		if (!IS_ERR(devlink_port))
 			info->user_ptr[1] = devlink_port;
+	} else if (ops->internal_flags & DEVLINK_NL_FLAG_NEED_RATE) {
+		struct devlink_rate *devlink_rate;
+
+		devlink_rate = devlink_rate_get_from_info(devlink, info);
+		if (IS_ERR(devlink_rate)) {
+			err = PTR_ERR(devlink_rate);
+			goto unlock;
+		}
+		info->user_ptr[1] = devlink_rate;
+	} else if (ops->internal_flags & DEVLINK_NL_FLAG_NEED_RATE_NODE) {
+		struct devlink_rate *rate_node;
+
+		rate_node = devlink_rate_node_get_from_info(devlink, info);
+		if (IS_ERR(rate_node)) {
+			err = PTR_ERR(rate_node);
+			goto unlock;
+		}
+		info->user_ptr[1] = rate_node;
 	}
 	return 0;
 
@@ -749,6 +843,56 @@ devlink_port_fn_hw_addr_fill(struct devlink *devlink, const struct devlink_ops *
 	return 0;
 }
 
+static int devlink_nl_rate_fill(struct sk_buff *msg,
+				struct devlink *devlink,
+				struct devlink_rate *devlink_rate,
+				enum devlink_command cmd, u32 portid,
+				u32 seq, int flags,
+				struct netlink_ext_ack *extack)
+{
+	void *hdr;
+
+	hdr = genlmsg_put(msg, portid, seq, &devlink_nl_family, flags, cmd);
+	if (!hdr)
+		return -EMSGSIZE;
+
+	if (devlink_nl_put_handle(msg, devlink))
+		goto nla_put_failure;
+
+	if (nla_put_u16(msg, DEVLINK_ATTR_RATE_TYPE, devlink_rate->type))
+		goto nla_put_failure;
+
+	if (devlink_rate_is_leaf(devlink_rate)) {
+		if (nla_put_u32(msg, DEVLINK_ATTR_PORT_INDEX,
+				devlink_rate->devlink_port->index))
+			goto nla_put_failure;
+	} else if (devlink_rate_is_node(devlink_rate)) {
+		if (nla_put_string(msg, DEVLINK_ATTR_RATE_NODE_NAME,
+				   devlink_rate->name))
+			goto nla_put_failure;
+	}
+
+	if (nla_put_u64_64bit(msg, DEVLINK_ATTR_RATE_TX_SHARE,
+			      devlink_rate->tx_share, DEVLINK_ATTR_PAD))
+		goto nla_put_failure;
+
+	if (nla_put_u64_64bit(msg, DEVLINK_ATTR_RATE_TX_MAX,
+			      devlink_rate->tx_max, DEVLINK_ATTR_PAD))
+		goto nla_put_failure;
+
+	if (devlink_rate->parent)
+		if (nla_put_string(msg, DEVLINK_ATTR_RATE_PARENT_NODE_NAME,
+				   devlink_rate->parent->name))
+			goto nla_put_failure;
+
+	genlmsg_end(msg, hdr);
+	return 0;
+
+nla_put_failure:
+	genlmsg_cancel(msg, hdr);
+	return -EMSGSIZE;
+}
+
 static bool
 devlink_port_fn_state_valid(enum devlink_port_fn_state state)
 {
@@ -920,6 +1064,111 @@ static void devlink_port_notify(struct devlink_port *devlink_port,
 				msg, 0, DEVLINK_MCGRP_CONFIG, GFP_KERNEL);
 }
 
+static void devlink_rate_notify(struct devlink_rate *devlink_rate,
+				enum devlink_command cmd)
+{
+	struct devlink *devlink = devlink_rate->devlink;
+	struct sk_buff *msg;
+	int err;
+
+	WARN_ON(cmd != DEVLINK_CMD_RATE_NEW &&
+		cmd != DEVLINK_CMD_RATE_DEL);
+
+	msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
+	if (!msg)
+		return;
+
+	err = devlink_nl_rate_fill(msg, devlink, devlink_rate,
+				   cmd, 0, 0, 0, NULL);
+	if (err) {
+		nlmsg_free(msg);
+		return;
+	}
+
+	genlmsg_multicast_netns(&devlink_nl_family, devlink_net(devlink),
+				msg, 0, DEVLINK_MCGRP_CONFIG, GFP_KERNEL);
+}
+
+static int devlink_nl_cmd_rate_get_dumpit(struct sk_buff *msg,
+					  struct netlink_callback *cb)
+{
+	struct devlink_rate *devlink_rate;
+	struct devlink *devlink;
+	int start = cb->args[0];
+	int idx = 0;
+	int err = 0;
+
+	mutex_lock(&devlink_mutex);
+	list_for_each_entry(devlink, &devlink_list, list) {
+		if (!net_eq(devlink_net(devlink), sock_net(msg->sk)))
+			continue;
+		mutex_lock(&devlink->lock);
+		list_for_each_entry(devlink_rate, &devlink->rate_list, list) {
+			enum devlink_command cmd = DEVLINK_CMD_RATE_NEW;
+			u32 id = NETLINK_CB(cb->skb).portid;
+
+			if (idx < start) {
+				idx++;
+				continue;
+			}
+			err = devlink_nl_rate_fill(msg, devlink,
+						   devlink_rate,
+						   cmd, id,
+						   cb->nlh->nlmsg_seq,
+						   NLM_F_MULTI, NULL);
+			if (err) {
+				mutex_unlock(&devlink->lock);
+				goto out;
+			}
+			idx++;
+		}
+		mutex_unlock(&devlink->lock);
+	}
+out:
+	mutex_unlock(&devlink_mutex);
+	if (err != -EMSGSIZE)
+		return err;
+
+	cb->args[0] = idx;
+	return msg->len;
+}
+
+static int devlink_nl_cmd_rate_get_doit(struct sk_buff *skb,
+					struct genl_info *info)
+{
+	struct devlink_rate *devlink_rate = info->user_ptr[1];
+	struct devlink *devlink = devlink_rate->devlink;
+	struct sk_buff *msg;
+	int err;
+
+	msg = nlmsg_new(NLMSG_DEFAULT_SIZE, GFP_KERNEL);
+	if (!msg)
+		return -ENOMEM;
+
+	err = devlink_nl_rate_fill(msg, devlink, devlink_rate,
+				   DEVLINK_CMD_RATE_NEW,
+				   info->snd_portid, info->snd_seq, 0,
+				   info->extack);
+	if (err) {
+		nlmsg_free(msg);
+		return err;
+	}
+
+	return genlmsg_reply(msg, info);
+}
+
+static bool
+devlink_rate_is_parent_node(struct devlink_rate *devlink_rate,
+			    struct devlink_rate *parent)
+{
+	while (parent) {
+		if (parent == devlink_rate)
+			return true;
+		parent = parent->parent;
+	}
+	return false;
+}
+
 static int devlink_nl_cmd_get_doit(struct sk_buff *skb, struct genl_info *info)
 {
 	struct devlink *devlink = info->user_ptr[0];
@@ -1340,6 +1589,255 @@ static int devlink_nl_cmd_port_del_doit(struct sk_buff *skb,
 	return devlink->ops->port_del(devlink, port_index, extack);
 }
 
+static int
+devlink_nl_rate_parent_node_set(struct devlink_rate *devlink_rate,
+				struct genl_info *info,
+				struct nlattr *nla_parent)
+{
+	struct devlink *devlink = devlink_rate->devlink;
+	const char *parent_name = nla_data(nla_parent);
+	const struct devlink_ops *ops = devlink->ops;
+	size_t len = strlen(parent_name);
+	struct devlink_rate *parent;
+	int err = -EOPNOTSUPP;
+
+	parent = devlink_rate->parent;
+	if (parent && len) {
+		NL_SET_ERR_MSG_MOD(info->extack, "Rate object already has parent.");
+		return -EBUSY;
+	} else if (parent && !len) {
+		if (devlink_rate_is_leaf(devlink_rate))
+			err = ops->rate_leaf_parent_set(devlink_rate, NULL,
+							devlink_rate->priv, NULL,
+							info->extack);
+		else if (devlink_rate_is_node(devlink_rate))
+			err = ops->rate_node_parent_set(devlink_rate, NULL,
+							devlink_rate->priv, NULL,
+							info->extack);
+		if (err)
+			return err;
+
+		refcount_dec(&parent->refcnt);
+		devlink_rate->parent = NULL;
+	} else if (!parent && len) {
+		parent = devlink_rate_node_get_by_name(devlink, parent_name);
+		if (IS_ERR(parent))
+			return -ENODEV;
+
+		if (parent == devlink_rate) {
+			NL_SET_ERR_MSG_MOD(info->extack, "Parent to self is not allowed");
+			return -EINVAL;
+		}
+
+		if (devlink_rate_is_node(devlink_rate) &&
+		    devlink_rate_is_parent_node(devlink_rate, parent->parent)) {
+			NL_SET_ERR_MSG_MOD(info->extack, "Node is already a parent of parent node.");
+			return -EEXIST;
+		}
+
+		if (devlink_rate_is_leaf(devlink_rate))
+			err = ops->rate_leaf_parent_set(devlink_rate, parent,
+							devlink_rate->priv, parent->priv,
+							info->extack);
+		else if (devlink_rate_is_node(devlink_rate))
+			err = ops->rate_node_parent_set(devlink_rate, parent,
+							devlink_rate->priv, parent->priv,
+							info->extack);
+		if (err)
+			return err;
+
+		refcount_inc(&parent->refcnt);
+		devlink_rate->parent = parent;
+	}
+
+	return 0;
+}
+
+static int devlink_nl_rate_set(struct devlink_rate *devlink_rate,
+			       const struct devlink_ops *ops,
+			       struct genl_info *info)
+{
+	struct nlattr *nla_parent, **attrs = info->attrs;
+	int err = -EOPNOTSUPP;
+	u64 rate;
+
+	if (attrs[DEVLINK_ATTR_RATE_TX_SHARE]) {
+		rate = nla_get_u64(attrs[DEVLINK_ATTR_RATE_TX_SHARE]);
+		if (devlink_rate_is_leaf(devlink_rate))
+			err = ops->rate_leaf_tx_share_set(devlink_rate, devlink_rate->priv,
+							  rate, info->extack);
+		else if (devlink_rate_is_node(devlink_rate))
+			err = ops->rate_node_tx_share_set(devlink_rate, devlink_rate->priv,
+							  rate, info->extack);
+		if (err)
+			return err;
+		devlink_rate->tx_share = rate;
+	}
+
+	if (attrs[DEVLINK_ATTR_RATE_TX_MAX]) {
+		rate = nla_get_u64(attrs[DEVLINK_ATTR_RATE_TX_MAX]);
+		if (devlink_rate_is_leaf(devlink_rate))
+			err = ops->rate_leaf_tx_max_set(devlink_rate, devlink_rate->priv,
+							rate, info->extack);
+		else if (devlink_rate_is_node(devlink_rate))
+			err = ops->rate_node_tx_max_set(devlink_rate, devlink_rate->priv,
+							rate, info->extack);
+		if (err)
+			return err;
+		devlink_rate->tx_max = rate;
+	}
+
+	nla_parent = attrs[DEVLINK_ATTR_RATE_PARENT_NODE_NAME];
+	if (nla_parent) {
+		err = devlink_nl_rate_parent_node_set(devlink_rate, info,
+						      nla_parent);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+static bool devlink_rate_set_ops_supported(const struct devlink_ops *ops,
+					   struct genl_info *info,
+					   enum devlink_rate_type type)
+{
+	struct nlattr **attrs = info->attrs;
+
+	if (type == DEVLINK_RATE_TYPE_LEAF) {
+		if (attrs[DEVLINK_ATTR_RATE_TX_SHARE] && !ops->rate_leaf_tx_share_set) {
+			NL_SET_ERR_MSG_MOD(info->extack, "TX share set isn't supported for the leafs");
+			return false;
+		}
+		if (attrs[DEVLINK_ATTR_RATE_TX_MAX] && !ops->rate_leaf_tx_max_set) {
+			NL_SET_ERR_MSG_MOD(info->extack, "TX max set isn't supported for the leafs");
+			return false;
+		}
+		if (attrs[DEVLINK_ATTR_RATE_PARENT_NODE_NAME] &&
+		    !ops->rate_leaf_parent_set) {
+			NL_SET_ERR_MSG_MOD(info->extack, "Parent set isn't supported for the leafs");
+			return false;
+		}
+	} else if (type == DEVLINK_RATE_TYPE_NODE) {
+		if (attrs[DEVLINK_ATTR_RATE_TX_SHARE] && !ops->rate_node_tx_share_set) {
+			NL_SET_ERR_MSG_MOD(info->extack, "TX share set isn't supported for the nodes");
+			return false;
+		}
+		if (attrs[DEVLINK_ATTR_RATE_TX_MAX] && !ops->rate_node_tx_max_set) {
+			NL_SET_ERR_MSG_MOD(info->extack, "TX max set isn't supported for the nodes");
+			return false;
+		}
+		if (attrs[DEVLINK_ATTR_RATE_PARENT_NODE_NAME] &&
+		    !ops->rate_node_parent_set) {
+			NL_SET_ERR_MSG_MOD(info->extack, "Parent set isn't supported for the nodes");
+			return false;
+		}
+	} else {
+		WARN_ON("Unknown type of rate object");
+		return false;
+	}
+
+	return true;
+}
+
+static int devlink_nl_cmd_rate_set_doit(struct sk_buff *skb,
+					struct genl_info *info)
+{
+	struct devlink_rate *devlink_rate = info->user_ptr[1];
+	struct devlink *devlink = devlink_rate->devlink;
+	const struct devlink_ops *ops = devlink->ops;
+	int err;
+
+	if (!ops || !devlink_rate_set_ops_supported(ops, info, devlink_rate->type))
+		return -EOPNOTSUPP;
+
+	err = devlink_nl_rate_set(devlink_rate, ops, info);
+
+	if (!err)
+		devlink_rate_notify(devlink_rate, DEVLINK_CMD_RATE_NEW);
+	return err;
+}
+
+static int devlink_nl_cmd_rate_new_doit(struct sk_buff *skb,
+					struct genl_info *info)
+{
+	struct devlink *devlink = info->user_ptr[0];
+	struct devlink_rate *rate_node;
+	const struct devlink_ops *ops;
+	int err;
+
+	ops = devlink->ops;
+	if (!ops || !ops->rate_node_new || !ops->rate_node_del) {
+		NL_SET_ERR_MSG_MOD(info->extack, "Rate nodes aren't supported");
+		return -EOPNOTSUPP;
+	}
+
+	if (!devlink_rate_set_ops_supported(ops, info, DEVLINK_RATE_TYPE_NODE))
+		return -EOPNOTSUPP;
+
+	rate_node = devlink_rate_node_get_from_attrs(devlink, info->attrs);
+	if (!IS_ERR(rate_node))
+		return -EEXIST;
+	else if (rate_node == ERR_PTR(-EINVAL))
+		return -EINVAL;
+
+	rate_node = kzalloc(sizeof(*rate_node), GFP_KERNEL);
+	if (!rate_node)
+		return -ENOMEM;
+
+	rate_node->devlink = devlink;
+	rate_node->type = DEVLINK_RATE_TYPE_NODE;
+	rate_node->name = nla_strdup(info->attrs[DEVLINK_ATTR_RATE_NODE_NAME], GFP_KERNEL);
+	if (!rate_node->name) {
+		err = -ENOMEM;
+		goto err_strdup;
+	}
+
+	err = ops->rate_node_new(rate_node, &rate_node->priv, info->extack);
+	if (err)
+		goto err_node_new;
+
+	err = devlink_nl_rate_set(rate_node, ops, info);
+	if (err)
+		goto err_rate_set;
+
+	refcount_set(&rate_node->refcnt, 1);
+	list_add(&rate_node->list, &devlink->rate_list);
+	devlink_rate_notify(rate_node, DEVLINK_CMD_RATE_NEW);
+	return 0;
+
+err_rate_set:
+	ops->rate_node_del(rate_node, rate_node->priv, info->extack);
+err_node_new:
+	kfree(rate_node->name);
+err_strdup:
+	kfree(rate_node);
+	return err;
+}
+
+static int devlink_nl_cmd_rate_del_doit(struct sk_buff *skb,
+					struct genl_info *info)
+{
+	struct devlink_rate *rate_node = info->user_ptr[1];
+	struct devlink *devlink = rate_node->devlink;
+	const struct devlink_ops *ops = devlink->ops;
+	int err;
+
+	if (refcount_read(&rate_node->refcnt) > 1) {
+		NL_SET_ERR_MSG_MOD(info->extack, "Node has children. Cannot delete node.");
+		return -EBUSY;
+	}
+
+	devlink_rate_notify(rate_node, DEVLINK_CMD_RATE_DEL);
+	err = ops->rate_node_del(rate_node, rate_node->priv, info->extack);
+	if (rate_node->parent)
+		refcount_dec(&rate_node->parent->refcnt);
+	list_del(&rate_node->list);
+	kfree(rate_node->name);
+	kfree(rate_node);
+	return err;
+}
+
 static int devlink_nl_sb_fill(struct sk_buff *msg, struct devlink *devlink,
 			      struct devlink_sb *devlink_sb,
 			      enum devlink_command cmd, u32 portid,
@@ -2208,6 +2706,30 @@ static int devlink_nl_cmd_eswitch_get_doit(struct sk_buff *skb,
 	return genlmsg_reply(msg, info);
 }
 
+static int devlink_rate_nodes_check(struct devlink *devlink, u16 mode,
+				    struct netlink_ext_ack *extack)
+{
+	struct devlink_rate *devlink_rate;
+	u16 old_mode;
+	int err;
+
+	if (!devlink->ops->eswitch_mode_get)
+		return -EOPNOTSUPP;
+	err = devlink->ops->eswitch_mode_get(devlink, &old_mode);
+	if (err)
+		return err;
+
+	if (old_mode == mode)
+		return 0;
+
+	list_for_each_entry(devlink_rate, &devlink->rate_list, list)
+		if (devlink_rate_is_node(devlink_rate)) {
+			NL_SET_ERR_MSG_MOD(extack, "Rate node(s) exists.");
+			return -EBUSY;
+		}
+	return 0;
+}
+
 static int devlink_nl_cmd_eswitch_set_doit(struct sk_buff *skb,
 					   struct genl_info *info)
 {
@@ -2222,6 +2744,9 @@ static int devlink_nl_cmd_eswitch_set_doit(struct sk_buff *skb,
 		if (!ops->eswitch_mode_set)
 			return -EOPNOTSUPP;
 		mode = nla_get_u16(info->attrs[DEVLINK_ATTR_ESWITCH_MODE]);
+		err = devlink_rate_nodes_check(devlink, mode, info->extack);
+		if (err)
+			return err;
 		err = ops->eswitch_mode_set(devlink, mode, info->extack);
 		if (err)
 			return err;
@@ -7802,6 +8327,11 @@ static const struct nla_policy devlink_nl_policy[DEVLINK_ATTR_MAX + 1] = {
 	[DEVLINK_ATTR_PORT_PCI_PF_NUMBER] = { .type = NLA_U16 },
 	[DEVLINK_ATTR_PORT_PCI_SF_NUMBER] = { .type = NLA_U32 },
 	[DEVLINK_ATTR_PORT_CONTROLLER_NUMBER] = { .type = NLA_U32 },
+	[DEVLINK_ATTR_RATE_TYPE] = { .type = NLA_U16 },
+	[DEVLINK_ATTR_RATE_TX_SHARE] = { .type = NLA_U64 },
+	[DEVLINK_ATTR_RATE_TX_MAX] = { .type = NLA_U64 },
+	[DEVLINK_ATTR_RATE_NODE_NAME] = { .type = NLA_NUL_STRING },
+	[DEVLINK_ATTR_RATE_PARENT_NODE_NAME] = { .type = NLA_NUL_STRING },
 };
 
 static const struct genl_small_ops devlink_nl_ops[] = {
@@ -7828,6 +8358,30 @@ static const struct genl_small_ops devlink_nl_ops[] = {
 		.internal_flags = DEVLINK_NL_FLAG_NEED_PORT,
 	},
 	{
+		.cmd = DEVLINK_CMD_RATE_GET,
+		.doit = devlink_nl_cmd_rate_get_doit,
+		.dumpit = devlink_nl_cmd_rate_get_dumpit,
+		.internal_flags = DEVLINK_NL_FLAG_NEED_RATE,
+		/* can be retrieved by unprivileged users */
+	},
+	{
+		.cmd = DEVLINK_CMD_RATE_SET,
+		.doit = devlink_nl_cmd_rate_set_doit,
+		.flags = GENL_ADMIN_PERM,
+		.internal_flags = DEVLINK_NL_FLAG_NEED_RATE,
+	},
+	{
+		.cmd = DEVLINK_CMD_RATE_NEW,
+		.doit = devlink_nl_cmd_rate_new_doit,
+		.flags = GENL_ADMIN_PERM,
+	},
+	{
+		.cmd = DEVLINK_CMD_RATE_DEL,
+		.doit = devlink_nl_cmd_rate_del_doit,
+		.flags = GENL_ADMIN_PERM,
+		.internal_flags = DEVLINK_NL_FLAG_NEED_RATE_NODE,
+	},
+	{
 		.cmd = DEVLINK_CMD_PORT_SPLIT,
 		.validate = GENL_DONT_VALIDATE_STRICT | GENL_DONT_VALIDATE_DUMP,
 		.doit = devlink_nl_cmd_port_split_doit,
@@ -8202,6 +8756,7 @@ struct devlink *devlink_alloc(const struct devlink_ops *ops, size_t priv_size)
 	xa_init_flags(&devlink->snapshot_ids, XA_FLAGS_ALLOC);
 	__devlink_net_set(devlink, &init_net);
 	INIT_LIST_HEAD(&devlink->port_list);
+	INIT_LIST_HEAD(&devlink->rate_list);
 	INIT_LIST_HEAD(&devlink->sb_list);
 	INIT_LIST_HEAD_RCU(&devlink->dpipe_table_list);
 	INIT_LIST_HEAD(&devlink->resource_list);
@@ -8304,6 +8859,7 @@ void devlink_free(struct devlink *devlink)
 	WARN_ON(!list_empty(&devlink->resource_list));
 	WARN_ON(!list_empty(&devlink->dpipe_table_list));
 	WARN_ON(!list_empty(&devlink->sb_list));
+	WARN_ON(!list_empty(&devlink->rate_list));
 	WARN_ON(!list_empty(&devlink->port_list));
 
 	xa_destroy(&devlink->snapshot_ids);
@@ -8620,6 +9176,108 @@ void devlink_port_attrs_pci_sf_set(struct devlink_port *devlink_port, u32 contro
 }
 EXPORT_SYMBOL_GPL(devlink_port_attrs_pci_sf_set);
 
+/**
+ * devlink_rate_leaf_create - create devlink rate leaf
+ *
+ * @devlink_port: devlink port object to create rate object on
+ * @priv: driver private data
+ *
+ * Create devlink rate object of type leaf on provided @devlink_port.
+ * Throws call trace if @devlink_port already has a devlink rate object.
+ *
+ * Context: Takes and release devlink->lock <mutex>.
+ *
+ * Return: -ENOMEM if failed to allocate rate object, 0 otherwise.
+ */
+int
+devlink_rate_leaf_create(struct devlink_port *devlink_port, void *priv)
+{
+	struct devlink *devlink = devlink_port->devlink;
+	struct devlink_rate *devlink_rate;
+
+	devlink_rate = kzalloc(sizeof(*devlink_rate), GFP_KERNEL);
+	if (!devlink_rate)
+		return -ENOMEM;
+
+	mutex_lock(&devlink->lock);
+	WARN_ON(devlink_port->devlink_rate);
+	devlink_rate->type = DEVLINK_RATE_TYPE_LEAF;
+	devlink_rate->devlink = devlink;
+	devlink_rate->devlink_port = devlink_port;
+	devlink_rate->priv = priv;
+	list_add_tail(&devlink_rate->list, &devlink->rate_list);
+	devlink_port->devlink_rate = devlink_rate;
+	devlink_rate_notify(devlink_rate, DEVLINK_CMD_RATE_NEW);
+	mutex_unlock(&devlink->lock);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(devlink_rate_leaf_create);
+
+/**
+ * devlink_rate_leaf_destroy - destroy devlink rate leaf
+ *
+ * @devlink_port: devlink port linked to the rate object
+ *
+ * Context: Takes and release devlink->lock <mutex>.
+ */
+void devlink_rate_leaf_destroy(struct devlink_port *devlink_port)
+{
+	struct devlink_rate *devlink_rate = devlink_port->devlink_rate;
+	struct devlink *devlink = devlink_port->devlink;
+
+	if (!devlink_rate)
+		return;
+
+	mutex_lock(&devlink->lock);
+	devlink_rate_notify(devlink_rate, DEVLINK_CMD_RATE_DEL);
+	list_del(&devlink_rate->list);
+	devlink_port->devlink_rate = NULL;
+	mutex_unlock(&devlink->lock);
+	kfree(devlink_rate);
+}
+EXPORT_SYMBOL_GPL(devlink_rate_leaf_destroy);
+
+/**
+ * devlink_rate_nodes_destroy - destroy all devlink rate nodes on device
+ *
+ * @devlink: devlink instance
+ *
+ * Unset parent for all rate objects and destroy all rate nodes
+ * on specified device.
+ *
+ * Context: Takes and release devlink->lock <mutex>.
+ */
+void devlink_rate_nodes_destroy(struct devlink *devlink)
+{
+	static struct devlink_rate *devlink_rate, *tmp;
+	const struct devlink_ops *ops = devlink->ops;
+
+	mutex_lock(&devlink->lock);
+	list_for_each_entry(devlink_rate, &devlink->rate_list, list) {
+		if (!devlink_rate->parent)
+			continue;
+
+		refcount_dec(&devlink_rate->parent->refcnt);
+		if (devlink_rate_is_leaf(devlink_rate))
+			ops->rate_leaf_parent_set(devlink_rate, NULL, devlink_rate->priv,
+						  NULL, NULL);
+		else if (devlink_rate_is_node(devlink_rate))
+			ops->rate_node_parent_set(devlink_rate, NULL, devlink_rate->priv,
+						  NULL, NULL);
+	}
+	list_for_each_entry_safe(devlink_rate, tmp, &devlink->rate_list, list) {
+		if (devlink_rate_is_node(devlink_rate)) {
+			ops->rate_node_del(devlink_rate, devlink_rate->priv, NULL);
+			list_del(&devlink_rate->list);
+			kfree(devlink_rate->name);
+			kfree(devlink_rate);
+		}
+	}
+	mutex_unlock(&devlink->lock);
+}
+EXPORT_SYMBOL_GPL(devlink_rate_nodes_destroy);
+
 static int __devlink_port_phys_port_name_get(struct devlink_port *devlink_port,
 					     char *name, size_t len)
 {