Merge branch 'net-introduce-tx-h-w-shaping-api'

Paolo Abeni says:

====================
net: introduce TX H/W shaping API

We have a plurality of shaping-related drivers API, but none flexible
enough to meet existing demand from vendors[1].

This series introduces new device APIs to configure in a flexible way
TX H/W shaping. The new functionalities are exposed via a newly
defined generic netlink interface and include introspection
capabilities. Some self-tests are included, on top of a dummy
netdevsim implementation. Finally a basic implementation for the iavf
driver is provided.

Some usage examples:

* Configure shaping on a given queue:

./tools/net/ynl/cli.py --spec Documentation/netlink/specs/shaper.yaml \
	--do set --json '{"ifindex": '$IFINDEX',
			  "shaper": {"handle":
				     {"scope": "queue", "id":'$QUEUEID'},
			  "bw-max": 2000000}}'

* Container B/W sharing

The orchestration infrastructure wants to group the
container-related queues under a RR scheduling and limit the aggregate
bandwidth:

./tools/net/ynl/cli.py --spec Documentation/netlink/specs/shaper.yaml \
	--do group --json '{"ifindex": '$IFINDEX',
			"leaves": [
			  {"handle": {"scope": "queue", "id":'$QID1'},
			   "weight": '$W1'},
			  {"handle": {"scope": "queue", "id":'$QID2'},
			   "weight": '$W2'}],
			  {"handle": {"scope": "queue", "id":'$QID3'},
			   "weight": '$W3'}],
			"handle": {"scope":"node"},
			"bw-max": 10000000}'
{'ifindex': $IFINDEX, 'handle': {'scope': 'node', 'id': 0}}

Q1 \
    \
Q2 -- node 0 -------  netdev
    / (bw-max: 10M)
Q3 /

* Delegation

A containers wants to limit the aggregate B/W bandwidth of 2 of the 3
queues it owns - the starting configuration is the one from the
previous point:

SPEC=Documentation/netlink/specs/net_shaper.yaml
./tools/net/ynl/cli.py --spec $SPEC \
	--do group --json '{"ifindex": '$IFINDEX',
			"leaves": [
			  {"handle": {"scope": "queue", "id":'$QID1'},
			   "weight": '$W1'},
			  {"handle": {"scope": "queue", "id":'$QID2'},
			   "weight": '$W2'}],
			"handle": {"scope": "node"},
			"bw-max": 5000000 }'
{'ifindex': $IFINDEX, 'handle': {'scope': 'node', 'id': 1}}

Q1 -- node 1 --------\
    / (bw-max: 5M)    \
Q2 /                   node 0 -------  netdev
                      /(bw-max: 10M)
Q3 ------------------/

In a group operation, when prior to the op itself, the leaves have
different parents, the user must specify the parent handle for the
group. I.e., starting from the previous config:

./tools/net/ynl/cli.py --spec $SPEC \
	--do group --json '{"ifindex": '$IFINDEX',
			"leaves": [
			  {"handle": {"scope": "queue", "id":'$QID1'},
			   "weight": '$W1'},
			  {"handle": {"scope": "queue", "id":'$QID3'},
			   "weight": '$W3'}],
			"handle": {"scope": "node"},
			"bw-max": 3000000 }'
Netlink error: Invalid argument
nl_len = 96 (80) nl_flags = 0x300 nl_type = 2
	error: -22
	extack: {'msg': 'All the leaves shapers must have the same old parent'}

./tools/net/ynl/cli.py --spec $SPEC \
	--do group --json '{"ifindex": '$IFINDEX',
			"leaves": [
			  {"handle": {"scope": "queue", "id":'$QID1'},
			   "weight": '$W1'},
			  {"handle": {"scope": "queue", "id":'$QID3'},
			   "weight": '$W3'}],
			"handle": {"scope": "node"},
			"parent": {"scope": "node", "id": 1},
			"bw-max": 3000000 }
{'ifindex': $IFINDEX, 'handle': {'scope': 'node', 'id': 2}}

Q1 -- node 2 ---
    /(bw-max:3M)\
Q3 /             \
         ---- node 1 \
        / (bw-max: 5M)\
      Q2              node 0 -------  netdev
                      (bw-max: 10M)

* Cleanup:

Still starting from config 1To delete a single queue shaper

./tools/net/ynl/cli.py --spec $SPEC --do delete --json \
	'{"ifindex": '$IFINDEX',
	  "handle": {"scope": "queue", "id":'$QID3'}}'

Q1 -- node 2 ---
     (bw-max:3M)\
                 \
         ---- node 1 \
        / (bw-max: 5M)\
      Q2              node 0 -------  netdev
                      (bw-max: 10M)

Deleting a node shaper relinks all its leaves to the node's parent:

./tools/net/ynl/cli.py --spec $SPEC --do delete --json \
	'{"ifindex": '$IFINDEX',
	  "handle": {"scope": "node", "id":2}}'

Q1 ---\
       \
        node 1----- \
       / (bw-max: 5M)\
Q2----/              node 0 -------  netdev
                     (bw-max: 10M)

Deleting the last shaper under a node shaper deletes the node, too:

./tools/net/ynl/cli.py --spec $SPEC --do delete --json \
	'{"ifindex": '$IFINDEX',
	  "handle": {"scope": "queue", "id":'$QID1'}}'
./tools/net/ynl/cli.py --spec $SPEC --do delete --json \
	'{"ifindex": '$IFINDEX',
	  "handle": {"scope": "queue", "id":'$QID2'}}'
./tools/net/ynl/cli.py --spec $SPEC --do get --json \
	'{"ifindex": '$IFINDEX',
	  "handle": {"scope": "node", "id": 1}}'
Netlink error: No such file or directory
nl_len = 44 (28) nl_flags = 0x300 nl_type = 2
	error: -2
	extack: {'bad-attr': '.handle'}

Such delete recurses on parents that are left over with no leaves:

./tools/net/ynl/cli.py --spec $SPEC --do get --json \
	'{"ifindex": '$IFINDEX',
	  "handle": {"scope": "node", "id": 0}}'
Netlink error: No such file or directory
nl_len = 44 (28) nl_flags = 0x300 nl_type = 2
	error: -2
	extack: {'bad-attr': '.handle'}

v8: https://lore.kernel.org/cover.1727704215.git.pabeni@redhat.com
v7: https://lore.kernel.org/cover.1725919039.git.pabeni@redhat.com
v6: https://lore.kernel.org/cover.1725457317.git.pabeni@redhat.com
v5: https://lore.kernel.org/cover.1724944116.git.pabeni@redhat.com
v4: https://lore.kernel.org/cover.1724165948.git.pabeni@redhat.com
v3: https://lore.kernel.org/cover.1722357745.git.pabeni@redhat.com
RFC v2: https://lore.kernel.org/cover.1721851988.git.pabeni@redhat.com
RFC v1: https://lore.kernel.org/cover.1719518113.git.pabeni@redhat.com
====================

Link: https://patch.msgid.link/cover.1728460186.git.pabeni@redhat.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

3 files changed, 143 insertions, 2 deletions

diff --git a/include/linux/avf/virtchnl.h b/include/linux/avf/virtchnl.h
index f41395264dca..223e433c39fe 100644
--- a/include/linux/avf/virtchnl.h
+++ b/include/linux/avf/virtchnl.h
@@ -89,6 +89,9 @@ enum virtchnl_rx_hsplit {
 	VIRTCHNL_RX_HSPLIT_SPLIT_SCTP    = 8,
 };
 
+enum virtchnl_bw_limit_type {
+	VIRTCHNL_BW_SHAPER = 0,
+};
 /* END GENERIC DEFINES */
 
 /* Opcodes for VF-PF communication. These are placed in the v_opcode field
@@ -151,6 +154,11 @@ enum virtchnl_ops {
 	VIRTCHNL_OP_DISABLE_VLAN_STRIPPING_V2 = 55,
 	VIRTCHNL_OP_ENABLE_VLAN_INSERTION_V2 = 56,
 	VIRTCHNL_OP_DISABLE_VLAN_INSERTION_V2 = 57,
+	/* opcode 57 - 65 are reserved */
+	VIRTCHNL_OP_GET_QOS_CAPS = 66,
+	/* opcode 68 through 111 are reserved */
+	VIRTCHNL_OP_CONFIG_QUEUE_BW = 112,
+	VIRTCHNL_OP_CONFIG_QUANTA = 113,
 	VIRTCHNL_OP_MAX,
 };
 
@@ -261,6 +269,7 @@ VIRTCHNL_CHECK_STRUCT_LEN(16, virtchnl_vsi_resource);
 #define VIRTCHNL_VF_OFFLOAD_RX_FLEX_DESC	BIT(26)
 #define VIRTCHNL_VF_OFFLOAD_ADV_RSS_PF		BIT(27)
 #define VIRTCHNL_VF_OFFLOAD_FDIR_PF		BIT(28)
+#define VIRTCHNL_VF_OFFLOAD_QOS			BIT(29)
 
 #define VF_BASE_MODE_OFFLOADS (VIRTCHNL_VF_OFFLOAD_L2 | \
 			       VIRTCHNL_VF_OFFLOAD_VLAN | \
@@ -1416,6 +1425,85 @@ struct virtchnl_fdir_del {
 
 VIRTCHNL_CHECK_STRUCT_LEN(12, virtchnl_fdir_del);
 
+struct virtchnl_shaper_bw {
+	/* Unit is Kbps */
+	u32 committed;
+	u32 peak;
+};
+
+VIRTCHNL_CHECK_STRUCT_LEN(8, virtchnl_shaper_bw);
+
+/* VIRTCHNL_OP_GET_QOS_CAPS
+ * VF sends this message to get its QoS Caps, such as
+ * TC number, Arbiter and Bandwidth.
+ */
+struct virtchnl_qos_cap_elem {
+	u8 tc_num;
+	u8 tc_prio;
+#define VIRTCHNL_ABITER_STRICT      0
+#define VIRTCHNL_ABITER_ETS         2
+	u8 arbiter;
+#define VIRTCHNL_STRICT_WEIGHT      1
+	u8 weight;
+	enum virtchnl_bw_limit_type type;
+	union {
+		struct virtchnl_shaper_bw shaper;
+		u8 pad2[32];
+	};
+};
+
+VIRTCHNL_CHECK_STRUCT_LEN(40, virtchnl_qos_cap_elem);
+
+struct virtchnl_qos_cap_list {
+	u16 vsi_id;
+	u16 num_elem;
+	struct virtchnl_qos_cap_elem cap[];
+};
+
+VIRTCHNL_CHECK_STRUCT_LEN(4, virtchnl_qos_cap_list);
+#define virtchnl_qos_cap_list_LEGACY_SIZEOF	44
+
+/* VIRTCHNL_OP_CONFIG_QUEUE_BW */
+struct virtchnl_queue_bw {
+	u16 queue_id;
+	u8 tc;
+	u8 pad;
+	struct virtchnl_shaper_bw shaper;
+};
+
+VIRTCHNL_CHECK_STRUCT_LEN(12, virtchnl_queue_bw);
+
+struct virtchnl_queues_bw_cfg {
+	u16 vsi_id;
+	u16 num_queues;
+	struct virtchnl_queue_bw cfg[];
+};
+
+VIRTCHNL_CHECK_STRUCT_LEN(4, virtchnl_queues_bw_cfg);
+#define virtchnl_queues_bw_cfg_LEGACY_SIZEOF	16
+
+enum virtchnl_queue_type {
+	VIRTCHNL_QUEUE_TYPE_TX			= 0,
+	VIRTCHNL_QUEUE_TYPE_RX			= 1,
+};
+
+/* structure to specify a chunk of contiguous queues */
+struct virtchnl_queue_chunk {
+	/* see enum virtchnl_queue_type */
+	s32 type;
+	u16 start_queue_id;
+	u16 num_queues;
+};
+
+VIRTCHNL_CHECK_STRUCT_LEN(8, virtchnl_queue_chunk);
+
+struct virtchnl_quanta_cfg {
+	u16 quanta_size;
+	struct virtchnl_queue_chunk queue_select;
+};
+
+VIRTCHNL_CHECK_STRUCT_LEN(12, virtchnl_quanta_cfg);
+
 #define __vss_byone(p, member, count, old)				      \
 	(struct_size(p, member, count) + (old - 1 - struct_size(p, member, 0)))
 
@@ -1438,6 +1526,8 @@ VIRTCHNL_CHECK_STRUCT_LEN(12, virtchnl_fdir_del);
 		 __vss(virtchnl_vlan_filter_list_v2, __vss_byelem, p, m, c),  \
 		 __vss(virtchnl_tc_info, __vss_byelem, p, m, c),	      \
 		 __vss(virtchnl_rdma_qvlist_info, __vss_byelem, p, m, c),     \
+		 __vss(virtchnl_qos_cap_list, __vss_byelem, p, m, c),	      \
+		 __vss(virtchnl_queues_bw_cfg, __vss_byelem, p, m, c),	      \
 		 __vss(virtchnl_rss_key, __vss_byone, p, m, c),		      \
 		 __vss(virtchnl_rss_lut, __vss_byone, p, m, c))
 
@@ -1637,6 +1727,35 @@ virtchnl_vc_validate_vf_msg(struct virtchnl_version_info *ver, u32 v_opcode,
 	case VIRTCHNL_OP_DISABLE_VLAN_INSERTION_V2:
 		valid_len = sizeof(struct virtchnl_vlan_setting);
 		break;
+	case VIRTCHNL_OP_GET_QOS_CAPS:
+		break;
+	case VIRTCHNL_OP_CONFIG_QUEUE_BW:
+		valid_len = virtchnl_queues_bw_cfg_LEGACY_SIZEOF;
+		if (msglen >= valid_len) {
+			struct virtchnl_queues_bw_cfg *q_bw =
+				(struct virtchnl_queues_bw_cfg *)msg;
+
+			valid_len = virtchnl_struct_size(q_bw, cfg,
+							 q_bw->num_queues);
+			if (q_bw->num_queues == 0) {
+				err_msg_format = true;
+				break;
+			}
+		}
+		break;
+	case VIRTCHNL_OP_CONFIG_QUANTA:
+		valid_len = sizeof(struct virtchnl_quanta_cfg);
+		if (msglen >= valid_len) {
+			struct virtchnl_quanta_cfg *q_quanta =
+				(struct virtchnl_quanta_cfg *)msg;
+
+			if (q_quanta->quanta_size == 0 ||
+			    q_quanta->queue_select.num_queues == 0) {
+				err_msg_format = true;
+				break;
+			}
+		}
+		break;
 	/* These are always errors coming from the VF. */
 	case VIRTCHNL_OP_EVENT:
 	case VIRTCHNL_OP_UNKNOWN:
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 3baf8e539b6f..e6b93d01e631 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -1603,6 +1603,14 @@ struct net_device_ops {
 	int			(*ndo_hwtstamp_set)(struct net_device *dev,
 						    struct kernel_hwtstamp_config *kernel_config,
 						    struct netlink_ext_ack *extack);
+
+#if IS_ENABLED(CONFIG_NET_SHAPER)
+	/**
+	 * @net_shaper_ops: Device shaping offload operations
+	 * see include/net/net_shapers.h
+	 */
+	const struct net_shaper_ops *net_shaper_ops;
+#endif
 };
 
 /**
@@ -2406,6 +2414,19 @@ struct net_device {
 
 	u64			max_pacing_offload_horizon;
 
+	/**
+	 * @lock: protects @net_shaper_hierarchy, feel free to use for other
+	 * netdev-scope protection. Ordering: take after rtnl_lock.
+	 */
+	struct mutex		lock;
+
+#if IS_ENABLED(CONFIG_NET_SHAPER)
+	/**
+	 * @net_shaper_hierarchy: data tracking the current shaper status
+	 *  see include/net/net_shapers.h
+	 */
+	struct net_shaper_hierarchy *net_shaper_hierarchy;
+#endif
 	u8			priv[] ____cacheline_aligned
 				       __counted_by(priv_len);
 } ____cacheline_aligned;
diff --git a/include/linux/netlink.h b/include/linux/netlink.h
index b332c2048c75..a3ca198a3a9e 100644
--- a/include/linux/netlink.h
+++ b/include/linux/netlink.h
@@ -34,6 +34,7 @@ struct netlink_skb_parms {
 
 #define NETLINK_CB(skb)		(*(struct netlink_skb_parms*)&((skb)->cb))
 #define NETLINK_CREDS(skb)	(&NETLINK_CB((skb)).creds)
+#define NETLINK_CTX_SIZE	48
 
 
 void netlink_table_grab(void);
@@ -293,7 +294,7 @@ struct netlink_callback {
 	int			flags;
 	bool			strict_check;
 	union {
-		u8		ctx[48];
+		u8		ctx[NETLINK_CTX_SIZE];
 
 		/* args is deprecated. Cast a struct over ctx instead
 		 * for proper type safety.
@@ -302,7 +303,7 @@ struct netlink_callback {
 	};
 };
 
-#define NL_ASSERT_DUMP_CTX_FITS(type_name)				\
+#define NL_ASSERT_CTX_FITS(type_name)					\
 	BUILD_BUG_ON(sizeof(type_name) >				\
 		     sizeof_field(struct netlink_callback, ctx))