Merge branch 'tc_flower_offload'

Amir Vadai says:

====================
sched,mlx5: Offloaded TC flower filter statistics

This patchset introduces counters support for offloaded cls_flower filters.
When the user calls 'tc show -s ..', fl_dump is called.
Before fl_dump() returns the statistics, it calls the NIC driver (using a new
ndo_setup_tc() command - TC_CLSFLOWER_STATS) to read the hardware counters and
update the statistics accordingly. A new TC action op was added (stats_update())
to be used by the NIC driver to update the statistics.

Patchset was applied and tested over commit ed7cbbc ("udp: Resolve NULL pointer
dereference over flow-based vxlan device")
====================

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

6 files changed, 150 insertions, 5 deletions

diff --git a/include/linux/mlx5/driver.h b/include/linux/mlx5/driver.h
index 9613143f0561..07b504f7eb84 100644
--- a/include/linux/mlx5/driver.h
+++ b/include/linux/mlx5/driver.h
@@ -41,6 +41,7 @@
 #include <linux/slab.h>
 #include <linux/vmalloc.h>
 #include <linux/radix-tree.h>
+#include <linux/workqueue.h>
 
 #include <linux/mlx5/device.h>
 #include <linux/mlx5/doorbell.h>
@@ -457,6 +458,17 @@ struct mlx5_irq_info {
 	char name[MLX5_MAX_IRQ_NAME];
 };
 
+struct mlx5_fc_stats {
+	struct list_head list;
+	struct list_head addlist;
+	/* protect addlist add/splice operations */
+	spinlock_t addlist_lock;
+
+	struct workqueue_struct *wq;
+	struct delayed_work work;
+	unsigned long next_query;
+};
+
 struct mlx5_eswitch;
 
 struct mlx5_priv {
@@ -520,6 +532,8 @@ struct mlx5_priv {
 	struct mlx5_flow_root_namespace *fdb_root_ns;
 	struct mlx5_flow_root_namespace *esw_egress_root_ns;
 	struct mlx5_flow_root_namespace *esw_ingress_root_ns;
+
+	struct mlx5_fc_stats		fc_stats;
 };
 
 enum mlx5_device_state {
diff --git a/include/linux/mlx5/fs.h b/include/linux/mlx5/fs.h
index 6467569ad76e..4b7a107d9c19 100644
--- a/include/linux/mlx5/fs.h
+++ b/include/linux/mlx5/fs.h
@@ -73,6 +73,7 @@ struct mlx5_flow_destination {
 		u32			tir_num;
 		struct mlx5_flow_table	*ft;
 		u32			vport_num;
+		struct mlx5_fc		*counter;
 	};
 };
 
@@ -125,4 +126,10 @@ void mlx5_del_flow_rule(struct mlx5_flow_rule *fr);
 int mlx5_modify_rule_destination(struct mlx5_flow_rule *rule,
 				 struct mlx5_flow_destination *dest);
 
+struct mlx5_fc *mlx5_flow_rule_counter(struct mlx5_flow_rule *rule);
+struct mlx5_fc *mlx5_fc_create(struct mlx5_core_dev *dev, bool aging);
+void mlx5_fc_destroy(struct mlx5_core_dev *dev, struct mlx5_fc *counter);
+void mlx5_fc_query_cached(struct mlx5_fc *counter,
+			  u64 *bytes, u64 *packets, u64 *lastuse);
+
 #endif
diff --git a/include/linux/mlx5/mlx5_ifc.h b/include/linux/mlx5/mlx5_ifc.h
index 4ce4ea422a10..9a05cd7e5890 100644
--- a/include/linux/mlx5/mlx5_ifc.h
+++ b/include/linux/mlx5/mlx5_ifc.h
@@ -202,6 +202,9 @@ enum {
 	MLX5_CMD_OP_SET_FLOW_TABLE_ENTRY          = 0x936,
 	MLX5_CMD_OP_QUERY_FLOW_TABLE_ENTRY        = 0x937,
 	MLX5_CMD_OP_DELETE_FLOW_TABLE_ENTRY       = 0x938,
+	MLX5_CMD_OP_ALLOC_FLOW_COUNTER            = 0x939,
+	MLX5_CMD_OP_DEALLOC_FLOW_COUNTER          = 0x93a,
+	MLX5_CMD_OP_QUERY_FLOW_COUNTER            = 0x93b,
 	MLX5_CMD_OP_MODIFY_FLOW_TABLE             = 0x93c
 };
 
@@ -265,7 +268,8 @@ struct mlx5_ifc_flow_table_fields_supported_bits {
 
 struct mlx5_ifc_flow_table_prop_layout_bits {
 	u8         ft_support[0x1];
-	u8         reserved_at_1[0x2];
+	u8         reserved_at_1[0x1];
+	u8         flow_counter[0x1];
 	u8	   flow_modify_en[0x1];
 	u8         modify_root[0x1];
 	u8         identified_miss_table_mode[0x1];
@@ -932,6 +936,8 @@ enum mlx5_flow_destination_type {
 	MLX5_FLOW_DESTINATION_TYPE_VPORT        = 0x0,
 	MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE   = 0x1,
 	MLX5_FLOW_DESTINATION_TYPE_TIR          = 0x2,
+
+	MLX5_FLOW_DESTINATION_TYPE_COUNTER      = 0x100,
 };
 
 struct mlx5_ifc_dest_format_struct_bits {
@@ -941,6 +947,19 @@ struct mlx5_ifc_dest_format_struct_bits {
 	u8         reserved_at_20[0x20];
 };
 
+struct mlx5_ifc_flow_counter_list_bits {
+	u8         reserved_at_0[0x10];
+	u8         flow_counter_id[0x10];
+
+	u8         reserved_at_20[0x20];
+};
+
+union mlx5_ifc_dest_format_struct_flow_counter_list_auto_bits {
+	struct mlx5_ifc_dest_format_struct_bits dest_format_struct;
+	struct mlx5_ifc_flow_counter_list_bits flow_counter_list;
+	u8         reserved_at_0[0x40];
+};
+
 struct mlx5_ifc_fte_match_param_bits {
 	struct mlx5_ifc_fte_match_set_lyr_2_4_bits outer_headers;
 
@@ -2006,6 +2025,7 @@ enum {
 	MLX5_FLOW_CONTEXT_ACTION_ALLOW     = 0x1,
 	MLX5_FLOW_CONTEXT_ACTION_DROP      = 0x2,
 	MLX5_FLOW_CONTEXT_ACTION_FWD_DEST  = 0x4,
+	MLX5_FLOW_CONTEXT_ACTION_COUNT     = 0x8,
 };
 
 struct mlx5_ifc_flow_context_bits {
@@ -2022,13 +2042,16 @@ struct mlx5_ifc_flow_context_bits {
 	u8         reserved_at_80[0x8];
 	u8         destination_list_size[0x18];
 
-	u8         reserved_at_a0[0x160];
+	u8         reserved_at_a0[0x8];
+	u8         flow_counter_list_size[0x18];
+
+	u8         reserved_at_c0[0x140];
 
 	struct mlx5_ifc_fte_match_param_bits match_value;
 
 	u8         reserved_at_1200[0x600];
 
-	struct mlx5_ifc_dest_format_struct_bits destination[0];
+	union mlx5_ifc_dest_format_struct_flow_counter_list_auto_bits destination[0];
 };
 
 enum {
@@ -3937,6 +3960,34 @@ struct mlx5_ifc_query_flow_group_in_bits {
 	u8         reserved_at_e0[0x120];
 };
 
+struct mlx5_ifc_query_flow_counter_out_bits {
+	u8         status[0x8];
+	u8         reserved_at_8[0x18];
+
+	u8         syndrome[0x20];
+
+	u8         reserved_at_40[0x40];
+
+	struct mlx5_ifc_traffic_counter_bits flow_statistics[0];
+};
+
+struct mlx5_ifc_query_flow_counter_in_bits {
+	u8         opcode[0x10];
+	u8         reserved_at_10[0x10];
+
+	u8         reserved_at_20[0x10];
+	u8         op_mod[0x10];
+
+	u8         reserved_at_40[0x80];
+
+	u8         clear[0x1];
+	u8         reserved_at_c1[0xf];
+	u8         num_of_counters[0x10];
+
+	u8         reserved_at_e0[0x10];
+	u8         flow_counter_id[0x10];
+};
+
 struct mlx5_ifc_query_esw_vport_context_out_bits {
 	u8         status[0x8];
 	u8         reserved_at_8[0x18];
@@ -5510,6 +5561,28 @@ struct mlx5_ifc_dealloc_pd_in_bits {
 	u8         reserved_at_60[0x20];
 };
 
+struct mlx5_ifc_dealloc_flow_counter_out_bits {
+	u8         status[0x8];
+	u8         reserved_at_8[0x18];
+
+	u8         syndrome[0x20];
+
+	u8         reserved_at_40[0x40];
+};
+
+struct mlx5_ifc_dealloc_flow_counter_in_bits {
+	u8         opcode[0x10];
+	u8         reserved_at_10[0x10];
+
+	u8         reserved_at_20[0x10];
+	u8         op_mod[0x10];
+
+	u8         reserved_at_40[0x10];
+	u8         flow_counter_id[0x10];
+
+	u8         reserved_at_60[0x20];
+};
+
 struct mlx5_ifc_create_xrc_srq_out_bits {
 	u8         status[0x8];
 	u8         reserved_at_8[0x18];
@@ -6237,6 +6310,28 @@ struct mlx5_ifc_alloc_pd_in_bits {
 	u8         reserved_at_40[0x40];
 };
 
+struct mlx5_ifc_alloc_flow_counter_out_bits {
+	u8         status[0x8];
+	u8         reserved_at_8[0x18];
+
+	u8         syndrome[0x20];
+
+	u8         reserved_at_40[0x10];
+	u8         flow_counter_id[0x10];
+
+	u8         reserved_at_60[0x20];
+};
+
+struct mlx5_ifc_alloc_flow_counter_in_bits {
+	u8         opcode[0x10];
+	u8         reserved_at_10[0x10];
+
+	u8         reserved_at_20[0x10];
+	u8         op_mod[0x10];
+
+	u8         reserved_at_40[0x40];
+};
+
 struct mlx5_ifc_add_vxlan_udp_dport_out_bits {
 	u8         status[0x8];
 	u8         reserved_at_8[0x18];
diff --git a/include/net/act_api.h b/include/net/act_api.h
index 03e322b30218..2cd9e9bb059a 100644
--- a/include/net/act_api.h
+++ b/include/net/act_api.h
@@ -106,6 +106,7 @@ struct tc_action_ops {
 			int bind);
 	int     (*walk)(struct net *, struct sk_buff *,
 			struct netlink_callback *, int, struct tc_action *);
+	void	(*stats_update)(struct tc_action *, u64, u32, u64);
 };
 
 struct tc_action_net {
@@ -178,10 +179,21 @@ int tcf_action_copy_stats(struct sk_buff *, struct tc_action *, int);
 
 #define tc_for_each_action(_a, _exts) \
 	list_for_each_entry(a, &(_exts)->actions, list)
+
+static inline void tcf_action_stats_update(struct tc_action *a, u64 bytes,
+					   u64 packets, u64 lastuse)
+{
+	if (!a->ops->stats_update)
+		return;
+
+	a->ops->stats_update(a, bytes, packets, lastuse);
+}
+
 #else /* CONFIG_NET_CLS_ACT */
 
 #define tc_no_actions(_exts) true
 #define tc_for_each_action(_a, _exts) while (0)
+#define tcf_action_stats_update(a, bytes, packets, lastuse)
 
 #endif /* CONFIG_NET_CLS_ACT */
 #endif
diff --git a/include/net/pkt_cls.h b/include/net/pkt_cls.h
index 8b4893878cf4..0f7efa88f210 100644
--- a/include/net/pkt_cls.h
+++ b/include/net/pkt_cls.h
@@ -426,6 +426,7 @@ static inline bool tc_flags_valid(u32 flags)
 enum tc_fl_command {
 	TC_CLSFLOWER_REPLACE,
 	TC_CLSFLOWER_DESTROY,
+	TC_CLSFLOWER_STATS,
 };
 
 struct tc_cls_flower_offload {
diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
index 46e55f0202a6..a1fd76c22a59 100644
--- a/include/net/sch_generic.h
+++ b/include/net/sch_generic.h
@@ -527,11 +527,27 @@ static inline bool qdisc_is_percpu_stats(const struct Qdisc *q)
 	return q->flags & TCQ_F_CPUSTATS;
 }
 
+static inline void _bstats_update(struct gnet_stats_basic_packed *bstats,
+				  __u64 bytes, __u32 packets)
+{
+	bstats->bytes += bytes;
+	bstats->packets += packets;
+}
+
 static inline void bstats_update(struct gnet_stats_basic_packed *bstats,
 				 const struct sk_buff *skb)
 {
-	bstats->bytes += qdisc_pkt_len(skb);
-	bstats->packets += skb_is_gso(skb) ? skb_shinfo(skb)->gso_segs : 1;
+	_bstats_update(bstats,
+		       qdisc_pkt_len(skb),
+		       skb_is_gso(skb) ? skb_shinfo(skb)->gso_segs : 1);
+}
+
+static inline void _bstats_cpu_update(struct gnet_stats_basic_cpu *bstats,
+				      __u64 bytes, __u32 packets)
+{
+	u64_stats_update_begin(&bstats->syncp);
+	_bstats_update(&bstats->bstats, bytes, packets);
+	u64_stats_update_end(&bstats->syncp);
 }
 
 static inline void bstats_cpu_update(struct gnet_stats_basic_cpu *bstats,