[PATCH v2 08/17] net/ntnic: add interfaces for flow API engine

[Serhii Iliushyk <sil-plv@napatech.com>]

Add ntnic basic flow filter functionality.

Signed-off-by: Serhii Iliushyk <sil-plv@napatech.com>
---
 drivers/net/ntnic/include/flow_api.h          |  89 ++
 drivers/net/ntnic/include/flow_api_actions.h  |  13 +
 drivers/net/ntnic/include/flow_api_engine.h   |  46 +
 drivers/net/ntnic/include/flow_filter.h       |  15 +
 drivers/net/ntnic/include/hw_mod_backend.h    |  70 ++
 drivers/net/ntnic/include/nt4ga_filter.h      |  16 +
 .../ntnic/include/stream_binary_flow_api.h    | 946 ++++++++++++++++++
 7 files changed, 1195 insertions(+)
 create mode 100644 drivers/net/ntnic/include/flow_api.h
 create mode 100644 drivers/net/ntnic/include/flow_api_actions.h
 create mode 100644 drivers/net/ntnic/include/flow_api_engine.h
 create mode 100644 drivers/net/ntnic/include/flow_filter.h
 create mode 100644 drivers/net/ntnic/include/hw_mod_backend.h
 create mode 100644 drivers/net/ntnic/include/nt4ga_filter.h
 create mode 100644 drivers/net/ntnic/include/stream_binary_flow_api.h

diff --git a/drivers/net/ntnic/include/flow_api.h b/drivers/net/ntnic/include/flow_api.h
new file mode 100644
index 0000000000..4ccdbcde14
--- /dev/null
+++ b/drivers/net/ntnic/include/flow_api.h
@@ -0,0 +1,89 @@
+/*
+ * SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2023 Napatech A/S
+ */
+
+#ifndef _FLOW_API_H_
+#define _FLOW_API_H_
+
+#include <pthread.h>
+#include <stdint.h>
+#include <assert.h>
+
+#include "ntlog.h"
+
+#include "flow_api_actions.h"
+#include "flow_api_engine.h"
+#include "hw_mod_backend.h"
+#include "stream_binary_flow_api.h"
+
+/*
+ * Flow NIC and Eth port device management
+ */
+
+struct hw_mod_resource_s {
+	uint8_t *alloc_bm;	/* allocation bitmap */
+	uint32_t *ref;	/* reference counter for each resource element */
+	uint32_t resource_count;/* number of total available entries */
+};
+
+struct flow_eth_dev {
+	struct flow_nic_dev *ndev;	/* NIC that owns this port device */
+	uint8_t port;	/* NIC port id */
+	uint32_t port_id;	/* App assigned port_id - may be DPDK port_id */
+
+	struct flow_queue_id_s rx_queue[FLOW_MAX_QUEUES + 1];	/* 0th for exception */
+	int num_queues;	/* VSWITCH has exceptions sent on queue 0 per design */
+
+	int rss_target_id;	/* QSL_HSH index if RSS needed QSL v6+ */
+	struct flow_eth_dev *next;
+};
+
+enum flow_nic_hash_e {
+	HASH_ALGO_ROUND_ROBIN = 0,
+	HASH_ALGO_5TUPLE,
+};
+
+/* registered NIC backends */
+struct flow_nic_dev {
+	uint8_t adapter_no;	/* physical adapter no in the host system */
+	uint16_t ports;	/* number of in-ports addressable on this NIC */
+	enum flow_eth_dev_profile
+	flow_profile;	/* flow profile this NIC is initially prepared for */
+	int flow_mgnt_prepared;
+
+	struct hw_mod_resource_s res[RES_COUNT];/* raw NIC resource allocation table */
+	void *km_res_handle;
+	void *kcc_res_handle;
+
+	void *flm_mtr_handle;
+	void *group_handle;
+	void *hw_db_handle;
+	void *id_table_handle;
+
+	/* statistics */
+	uint32_t flow_stat_id_map[MAX_COLOR_FLOW_STATS];
+
+	uint32_t flow_unique_id_counter;
+	/* linked list of all flows created on this NIC */
+	struct flow_handle *flow_base;
+	/* linked list of all FLM flows created on this NIC */
+	struct flow_handle *flow_base_flm;
+	pthread_mutex_t flow_mtx;
+
+	/* NIC backend API */
+	struct flow_api_backend_s be;
+	/* linked list of created eth-port devices on this NIC */
+	struct flow_eth_dev *eth_base;
+	pthread_mutex_t mtx;
+
+	/* pre allocated default QSL Drop */
+	int default_qsl_drop_index;
+	/* pre allocated default QSL Discard */
+	int default_qsl_discard_index;
+
+	/* next NIC linked list */
+	struct flow_nic_dev *next;
+};
+
+#endif
diff --git a/drivers/net/ntnic/include/flow_api_actions.h b/drivers/net/ntnic/include/flow_api_actions.h
new file mode 100644
index 0000000000..f62cda5dc5
--- /dev/null
+++ b/drivers/net/ntnic/include/flow_api_actions.h
@@ -0,0 +1,13 @@
+/*
+ * SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2023 Napatech A/S
+ */
+
+#ifndef _FLOW_ACTIONS_H_
+#define _FLOW_ACTIONS_H_
+
+#include <stdint.h>
+
+#define MAX_COLOR_FLOW_STATS 0x400
+
+#endif	/* _FLOW_ACTIONS_H_ */
diff --git a/drivers/net/ntnic/include/flow_api_engine.h b/drivers/net/ntnic/include/flow_api_engine.h
new file mode 100644
index 0000000000..2fb43beac0
--- /dev/null
+++ b/drivers/net/ntnic/include/flow_api_engine.h
@@ -0,0 +1,46 @@
+/*
+ * SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2023 Napatech A/S
+ */
+
+#ifndef _FLOW_API_ENGINE_H_
+#define _FLOW_API_ENGINE_H_
+
+#include <stdatomic.h>
+#include <stdint.h>
+
+#include "hw_mod_backend.h"
+#include "stream_binary_flow_api.h"
+
+/*
+ * Resource management
+ * These are free resources in FPGA
+ * Other FPGA memory lists are linked to one of these
+ * and will implicitly follow them
+ */
+enum res_type_e {
+	RES_QUEUE,
+	RES_CAT_CFN,
+	RES_CAT_COT,
+	RES_CAT_EXO,
+	RES_CAT_LEN,
+	RES_KM_FLOW_TYPE,
+	RES_KM_CATEGORY,
+	RES_HSH_RCP,
+	RES_PDB_RCP,
+	RES_QSL_RCP,
+	RES_QSL_QST,
+	RES_SLC_RCP,
+	RES_SLC_LR_RCP,
+	RES_IOA_RCP,
+	RES_ROA_RCP,
+	RES_FLM_FLOW_TYPE,
+	RES_FLM_RCP,
+	RES_TPE_RCP,
+	RES_TPE_EXT,
+	RES_TPE_RPL,
+	RES_COUNT,
+	RES_INVALID
+};
+
+#endif	/* _FLOW_API_ENGINE_H_ */
diff --git a/drivers/net/ntnic/include/flow_filter.h b/drivers/net/ntnic/include/flow_filter.h
new file mode 100644
index 0000000000..53f77a9b60
--- /dev/null
+++ b/drivers/net/ntnic/include/flow_filter.h
@@ -0,0 +1,15 @@
+/*
+ * SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2023 Napatech A/S
+ */
+
+#ifndef __FLOW_FILTER_HPP__
+#define __FLOW_FILTER_HPP__
+
+#include "flow_api.h"
+#include "nthw_fpga_model.h"
+
+int flow_filter_init(nthw_fpga_t *p_fpga, struct flow_nic_dev **p_flow_device, int adapter_no);
+int flow_filter_done(struct flow_nic_dev *dev);
+
+#endif	/* __FLOW_FILTER_HPP__ */
diff --git a/drivers/net/ntnic/include/hw_mod_backend.h b/drivers/net/ntnic/include/hw_mod_backend.h
new file mode 100644
index 0000000000..ab3ce1f469
--- /dev/null
+++ b/drivers/net/ntnic/include/hw_mod_backend.h
@@ -0,0 +1,70 @@
+/*
+ * SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2023 Napatech A/S
+ */
+
+#ifndef _HW_MOD_BACKEND_H_
+#define _HW_MOD_BACKEND_H_
+
+#include <stdbool.h>
+#include <stdint.h>
+#include <string.h>
+
+#include "ntlog.h"
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#define COMMON_FUNC_INFO_S                                                                        \
+	int ver;                                                                                  \
+	void *base;                                                                               \
+	unsigned int alloced_size;                                                                \
+	int debug
+
+struct common_func_s {
+	COMMON_FUNC_INFO_S;
+};
+
+struct flm_func_s {
+	COMMON_FUNC_INFO_S;
+	uint32_t nb_categories;
+	uint32_t nb_size_mb;
+	uint32_t nb_entry_size;
+	uint32_t nb_variant;
+	uint32_t nb_prios;
+	uint32_t nb_pst_profiles;
+	uint32_t nb_scrub_profiles;
+	uint32_t nb_rpp_clock_in_ps;
+	uint32_t nb_load_aps_max;
+};
+
+struct tpe_func_s {
+	COMMON_FUNC_INFO_S;
+	uint32_t nb_rcp_categories;
+	uint32_t nb_ifr_categories;
+	uint32_t nb_cpy_writers;
+	uint32_t nb_rpl_depth;
+	uint32_t nb_rpl_ext_categories;
+};
+
+struct flow_api_backend_s {
+	void *be_dev;
+	/* flow filter FPGA modules */
+	struct flm_func_s flm;
+	struct tpe_func_s tpe;
+
+	/* NIC attributes */
+	unsigned int num_phy_ports;
+	unsigned int num_rx_ports;
+
+	/* flow filter resource capacities */
+	unsigned int max_categories;
+	unsigned int max_queues;
+};
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif	/* _HW_MOD_BACKEND_H_ */
diff --git a/drivers/net/ntnic/include/nt4ga_filter.h b/drivers/net/ntnic/include/nt4ga_filter.h
new file mode 100644
index 0000000000..8d1abbd2ee
--- /dev/null
+++ b/drivers/net/ntnic/include/nt4ga_filter.h
@@ -0,0 +1,16 @@
+/*
+ * SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2023 Napatech A/S
+ */
+
+#ifndef NT4GA_FILTER_H_
+#define NT4GA_FILTER_H_
+
+typedef struct nt4ga_filter_s {
+	int n_intf_cnt;
+	int n_queues_per_intf_cnt;
+
+	struct flow_nic_dev *mp_flow_device;
+} nt4ga_filter_t;
+
+#endif	/* NT4GA_FILTER_H_ */
diff --git a/drivers/net/ntnic/include/stream_binary_flow_api.h b/drivers/net/ntnic/include/stream_binary_flow_api.h
new file mode 100644
index 0000000000..096a349cc0
--- /dev/null
+++ b/drivers/net/ntnic/include/stream_binary_flow_api.h
@@ -0,0 +1,946 @@
+/*
+ * SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2023 Napatech A/S
+ */
+
+#ifndef _STREAM_BINARY_FLOW_API_H_
+#define _STREAM_BINARY_FLOW_API_H_
+
+#include <stdint.h>	/* uint16_t, uint32_t, uint64_t */
+#include <stdio.h>	/* snprintf */
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+typedef uint16_t be16_t;/* 16-bit big-endian */
+typedef uint32_t be32_t;/* 32-bit big-endian */
+typedef uint64_t be64_t;/* 64-bit big-endian */
+
+/* Max length for socket name, interface name, etc. */
+#define MAX_PATH_LEN 128
+
+/* Max RSS hash key length in bytes */
+#define MAX_RSS_KEY_LEN 40
+
+/** NT specific MASKs for RSS configuration **/
+#define NT_ETH_RSS_IPV4_MASK                                                                      \
+	(RTE_ETH_RSS_IPV4 | RTE_ETH_RSS_FRAG_IPV4 | RTE_ETH_RSS_NONFRAG_IPV4_OTHER |              \
+	 RTE_ETH_RSS_NONFRAG_IPV4_SCTP | RTE_ETH_RSS_NONFRAG_IPV4_TCP |                           \
+	 RTE_ETH_RSS_NONFRAG_IPV4_UDP)
+
+#define NT_ETH_RSS_IPV6_MASK                                                                      \
+	(RTE_ETH_RSS_IPV6 | RTE_ETH_RSS_FRAG_IPV6 | RTE_ETH_RSS_IPV6_EX |                         \
+	 RTE_ETH_RSS_IPV6_TCP_EX | RTE_ETH_RSS_IPV6_UDP_EX | RTE_ETH_RSS_NONFRAG_IPV6_OTHER |     \
+	 RTE_ETH_RSS_NONFRAG_IPV6_SCTP | RTE_ETH_RSS_NONFRAG_IPV6_TCP |                           \
+	 RTE_ETH_RSS_NONFRAG_IPV6_UDP)
+
+#define NT_ETH_RSS_IP_MASK                                                                        \
+	(NT_ETH_RSS_IPV4_MASK | NT_ETH_RSS_IPV6_MASK | RTE_ETH_RSS_L3_SRC_ONLY |                  \
+	 RTE_ETH_RSS_L3_DST_ONLY)
+
+/* List of all RSS flags supported for RSS calculation offload */
+#define NT_ETH_RSS_OFFLOAD_MASK                                                                   \
+	(RTE_ETH_RSS_ETH | RTE_ETH_RSS_L2_PAYLOAD | RTE_ETH_RSS_IP | RTE_ETH_RSS_TCP |            \
+	 RTE_ETH_RSS_UDP | RTE_ETH_RSS_SCTP | RTE_ETH_RSS_L2_SRC_ONLY | RTE_ETH_RSS_L2_DST_ONLY | \
+	 RTE_ETH_RSS_L4_SRC_ONLY | RTE_ETH_RSS_L4_DST_ONLY | RTE_ETH_RSS_L3_SRC_ONLY |            \
+	 RTE_ETH_RSS_L3_DST_ONLY | RTE_ETH_RSS_VLAN | RTE_ETH_RSS_LEVEL_MASK |                    \
+	 RTE_ETH_RSS_IPV4_CHKSUM | RTE_ETH_RSS_L4_CHKSUM | RTE_ETH_RSS_PORT | RTE_ETH_RSS_GTPU)
+
+/*
+ * Flow frontend for binary programming interface
+ */
+
+#define FLOW_MAX_QUEUES 128
+
+#define RAW_ENCAP_DECAP_ELEMS_MAX 16
+
+/*
+ * Partial flow mark and special flow marks
+ */
+#define FLOW_MARK_LACP 0x7fffffff
+#define FLOW_MARK_MAX 0x7ffffffe
+/*
+ * Flow eth dev profile determines how the FPGA module resources are
+ * managed and what features are available
+ */
+enum flow_eth_dev_profile {
+	FLOW_ETH_DEV_PROFILE_VSWITCH = 0,
+	FLOW_ETH_DEV_PROFILE_INLINE = 1,
+};
+
+/*
+ * Flow rule attributes
+ */
+struct flow_attr {
+	uint32_t group;	/* Priority group. */
+	uint32_t priority;	/* Rule priority level within group. */
+	uint16_t forced_vlan_vid;	/* Forced VLAN VID that filter must match. Ignored if 0. */
+	uint16_t caller_id;	/* Unique ID of caller application. */
+};
+
+struct flow_queue_id_s {
+	int id;
+	int hw_id;
+};
+
+/* NT Private rte flow items. */
+
+/* NT Private rte flow actions. */
+
+enum flow_elem_type {
+	FLOW_ELEM_TYPE_END,
+	FLOW_ELEM_TYPE_ANY,
+	FLOW_ELEM_TYPE_ETH,
+	FLOW_ELEM_TYPE_VLAN,
+	FLOW_ELEM_TYPE_IPV4,
+	FLOW_ELEM_TYPE_IPV6,
+	FLOW_ELEM_TYPE_SCTP,
+	FLOW_ELEM_TYPE_TCP,
+	FLOW_ELEM_TYPE_UDP,
+	FLOW_ELEM_TYPE_ICMP,
+	FLOW_ELEM_TYPE_ICMP6,
+	FLOW_ELEM_TYPE_VXLAN,
+	FLOW_ELEM_TYPE_GTP,
+	FLOW_ELEM_TYPE_GTP_PSC,
+	FLOW_ELEM_TYPE_PORT_ID,
+	FLOW_ELEM_TYPE_TAG,
+	FLOW_ELEM_TYPE_VOID,
+
+	/*
+	 * not associated with a RTE_ITEM..., but rather
+	 * an restoration API device specific extension
+	 */
+	FLOW_ELEM_TYPE_TUNNEL
+};
+
+enum flow_action_type {	/* conf structure */
+	FLOW_ACTION_TYPE_END,	/* -none- : End tag for action list */
+	FLOW_ACTION_TYPE_POP_VLAN,	/* -none- : Pops outer vlan tag */
+	FLOW_ACTION_TYPE_PUSH_VLAN,	/* struct flow_action_push_vlan : Push VLAN TAG */
+	FLOW_ACTION_TYPE_SET_VLAN_VID,	/* struct flow_action_set_vlan_vid : Set VLAN VID */
+	FLOW_ACTION_TYPE_SET_VLAN_PCP,	/* struct flow_action_set_vlan_pcp : Set VLAN PCP */
+	/* -none- : Decapsulate outer most VXLAN tunnel from matched flow */
+	FLOW_ACTION_TYPE_VXLAN_DECAP,
+	FLOW_ACTION_TYPE_VXLAN_ENCAP,	/* struct flow_action_vxlan_encap */
+	FLOW_ACTION_TYPE_DROP,	/* -none- : Drop packets of this flow */
+	FLOW_ACTION_TYPE_COUNT,	/* struct flow_action_count : Used for "query" flow function */
+	/* struct flow_action_mark : Used to tag a flow in HW with a MARK */
+	FLOW_ACTION_TYPE_MARK,
+	/* struct flow_action_tag : Used to tag a flow in HW with a TAG */
+	FLOW_ACTION_TYPE_SET_TAG,
+	/* struct flow_action_port_id : Destination port ID - HW port ID */
+	FLOW_ACTION_TYPE_PORT_ID,
+	FLOW_ACTION_TYPE_RSS,	/* struct flow_action_rss : */
+	FLOW_ACTION_TYPE_QUEUE,	/* struct flow_action_queue : */
+	FLOW_ACTION_TYPE_JUMP,	/* struct flow_action_jump : */
+	/* struct flow_action_meter : Used to set MBR record ids in FLM learn records */
+	FLOW_ACTION_TYPE_METER,
+	FLOW_ACTION_TYPE_RAW_ENCAP,	/* struct flow_action_raw_encap : */
+	FLOW_ACTION_TYPE_RAW_DECAP,	/* struct flow_action_raw_decap : */
+	FLOW_ACTION_TYPE_MODIFY_FIELD,	/* struct flow_action_modify_field : */
+
+	/*
+	 * -none- : not associated with a RTE_ACTION...,
+	 * but rather an restoration API device specific extension
+	 */
+	FLOW_ACTION_TYPE_TUNNEL_SET
+};
+
+#pragma pack(1)
+struct ether_addr_s {
+	uint8_t addr_b[6];
+};
+#pragma pack()
+
+static inline void flow_ether_format_addr(char *buf, uint16_t size,
+	const struct ether_addr_s *eth_addr)
+{
+	snprintf(buf, size, "%02X:%02X:%02X:%02X:%02X:%02X", eth_addr
+		->addr_b[0],
+		eth_addr
+		->addr_b[1], eth_addr
+		->addr_b[2], eth_addr
+		->addr_b[3],
+		eth_addr
+		->addr_b[4], eth_addr
+		->addr_b[5]);
+}
+
+/*
+ * IPv4 Header
+ */
+#pragma pack(1)
+struct ipv4_hdr_s {
+	uint8_t version_ihl;
+	uint8_t tos;
+	be16_t length;
+	be16_t id;
+	be16_t frag_offset;
+	uint8_t ttl;
+	uint8_t next_proto_id;
+	be16_t hdr_csum;
+	be32_t src_ip;
+	be32_t dst_ip;
+};
+#pragma pack()
+/*
+ * IPv6 Header
+ */
+#pragma pack(1)
+struct ipv6_hdr_s {
+	be32_t vtc_flow;/* IP version, traffic class & flow label */
+	be16_t payload_len;	/* IP packet length - includes ip header */
+	uint8_t proto;
+	uint8_t hop_limits;
+	uint8_t src_addr[16];
+	uint8_t dst_addr[16];
+};
+#pragma pack()
+
+/*
+ * SCTP Header
+ */
+#pragma pack(1)
+struct sctp_hdr_s {
+	be16_t src_port;
+	be16_t dst_port;
+	be32_t tag;	/* Validation tag */
+	be32_t cksum;
+};
+#pragma pack()
+
+/*
+ * TCP Header
+ */
+#pragma pack(1)
+struct tcp_hdr_s {
+	be16_t src_port;
+	be16_t dst_port;
+	be32_t sent_seq;
+	be32_t recv_ack;
+	uint8_t data_off;
+	uint8_t tcp_flags;
+	be16_t rx_win;
+	be16_t cksum;
+	be16_t tcp_urp;
+};
+#pragma pack()
+
+/*
+ * UDP Header
+ */
+#pragma pack(1)
+struct udp_hdr_s {
+	be16_t src_port;
+	be16_t dst_port;
+	be16_t len;
+	be16_t cksum;
+};
+#pragma pack()
+
+/*
+ * ICMP Header
+ */
+#pragma pack(1)
+struct icmp_hdr_s {
+	uint8_t type;
+	uint8_t code;
+	be16_t cksum;
+	be16_t ident;
+	be16_t seq_nb;
+};
+#pragma pack()
+/*
+ * FLOW_ELEM_TYPE_ETH specification
+ */
+#pragma pack(1)
+struct flow_elem_eth {
+	struct ether_addr_s d_addr;	/* DMAC */
+	struct ether_addr_s s_addr;	/* SMAC */
+	be16_t ether_type;	/* Frame type */
+};
+#pragma pack()
+
+/*
+ * FLOW_ELEM_TYPE_VLAN specification
+ */
+#pragma pack(1)
+struct flow_elem_vlan {
+	be16_t tci;	/* Tag control information */
+	be16_t inner_type;	/* Inner EtherType or TPID */
+};
+#pragma pack()
+
+/*
+ * FLOW_ELEM_TYPE_IPV4 specification
+ */
+struct flow_elem_ipv4 {
+	struct ipv4_hdr_s hdr;
+};
+
+/*
+ * FLOW_ELEM_TYPE_IPV6 specification
+ */
+struct flow_elem_ipv6 {
+	struct ipv6_hdr_s hdr;
+};
+
+/*
+ * FLOW_ELEM_TYPE_SCTP specification
+ */
+struct flow_elem_sctp {
+	struct sctp_hdr_s hdr;
+};
+
+/*
+ * FLOW_ELEM_TYPE_TCP specification
+ */
+struct flow_elem_tcp {
+	struct tcp_hdr_s hdr;
+};
+
+/*
+ * FLOW_ELEM_TYPE_UDP specification
+ */
+struct flow_elem_udp {
+	struct udp_hdr_s hdr;
+};
+
+/*
+ * FLOW_ELEM_TYPE_ICMP specification
+ */
+struct flow_elem_icmp {
+	struct icmp_hdr_s hdr;
+};
+
+/*
+ * FLOW_ELEM_TYPE_ICMP6 specification
+ */
+#pragma pack(1)
+struct flow_elem_icmp6 {
+	uint8_t type;	/**< ICMPv6 type. */
+	uint8_t code;	/**< ICMPv6 code. */
+	be16_t checksum;/**< ICMPv6 checksum. */
+};
+#pragma pack()
+
+/*
+ * FLOW_ELEM_TYPE_GTP specification
+ */
+#pragma pack(1)
+struct flow_elem_gtp {
+	uint8_t v_pt_rsv_flags;
+	uint8_t msg_type;
+	be16_t msg_len;
+	be32_t teid;
+};
+#pragma pack()
+
+/*
+ * FLOW_ELEM_TYPE_GTP_PSC specification
+ */
+#pragma pack(1)
+struct flow_elem_gtp_psc {
+	uint8_t hdr_len;
+	uint8_t pdu_type;
+	uint8_t qfi;
+};
+#pragma pack()
+
+/*
+ * FLOW_ELEM_TYPE_VXLAN specification (RFC 7348)
+ */
+#pragma pack(1)
+struct flow_elem_vxlan {
+	uint8_t flags;	/* Normally 0x08 (I flag) */
+	uint8_t rsvd0[3];
+	uint8_t vni[3];
+	uint8_t rsvd1;
+};
+#pragma pack()
+/*
+ * FLOW_ELEM_TYPE_PORT_ID specification
+ */
+struct flow_elem_port_id {
+	uint32_t id;	/* HW port no */
+};
+
+/*
+ * FLOW_ELEM_TYPE_TAG specification
+ */
+struct flow_elem_tag {
+	uint32_t data;
+	uint8_t index;
+};
+
+/*
+ * FLOW_ELEM_TYPE_ANY specification
+ */
+struct flow_elem_any {
+	uint32_t num;	/* *< Number of layers covered. */
+};
+
+struct flow_elem {
+	enum flow_elem_type type;	/* element type */
+	const void *spec;	/* Pointer to element specification structure */
+	const void *mask;	/* Bitmask applied to spec - same type */
+};
+
+/* Note: Keep in sync with the rte_eth_hash_function structure defined in rte_ethdev.h */
+enum nt_eth_hash_function {
+	NT_ETH_HASH_FUNCTION_DEFAULT = 0,
+	NT_ETH_HASH_FUNCTION_TOEPLITZ,
+	NT_ETH_HASH_FUNCTION_SIMPLE_XOR,
+	NT_ETH_HASH_FUNCTION_SYMMETRIC_TOEPLITZ,
+	NT_ETH_HASH_FUNCTION_SYMMETRIC_TOEPLITZ_SORT,
+	NT_ETH_HASH_FUNCTION_MAX,
+};
+
+struct flow_action_rss {
+	enum nt_eth_hash_function func;
+	/*
+	 * UNUSED; rte_flow_action_rss->level is used to set
+	 * RTE_ETH_RSS_LEVEL_OUTERMOST & RTE_ETH_RSS_LEVEL_INNERMOST
+	 * bits at 'flow_action_rss->types' below
+	 */
+	uint32_t level;
+	uint64_t types;	/* Specific RSS hash types (see like RTE_ETH_RSS_*) */
+	uint32_t key_len;	/* Hash key length in bytes supported for Toeplitz */
+	uint32_t queue_num;	/* Number of entries in queue */
+	const uint8_t *key;	/* Hash key supported for Toeplitz */
+	const uint16_t *queue;	/* Queue indices to use */
+};
+
+/*
+ * FLOW_ACTION_TYPE_PUSH_VLAN
+ * Push a new vlan TAG
+ */
+struct flow_action_push_vlan {
+	be16_t ethertype;
+};
+
+/*
+ * FLOW_ACTION_TYPE_SET_VLAN_VID
+ */
+struct flow_action_set_vlan_vid {
+	be16_t vlan_vid;
+};
+
+/*
+ * FLOW_ACTION_TYPE_SET_VLAN_PCP
+ */
+struct flow_action_set_vlan_pcp {
+	uint8_t vlan_pcp;	/* *< VLAN priority. */
+};
+
+/*
+ * FLOW_ACTION_TYPE_VXLAN_ENCAP specification
+ * Valid flow definition:
+ *
+ * - ETH / IPV4 / UDP / VXLAN / END
+ * - ETH / IPV6 / UDP / VXLAN / END
+ * - ETH / VLAN / IPV4 / UDP / VXLAN / END
+ */
+struct flow_action_vxlan_encap {
+	/* Encapsulating vxlan tunnel definition */
+	struct flow_elem *vxlan_tunnel;
+};
+
+/*
+ * FLOW_ACTION_TYPE_COUNT specification
+ */
+struct flow_action_count {
+	uint32_t id;	/* HW port no */
+};
+
+/*
+ * FLOW_ACTION_TYPE_COUNT specification (query)
+ */
+struct flow_query_count {
+	uint32_t reset : 1;
+	uint32_t hits_set : 1;
+	uint32_t bytes_set : 1;
+
+	uint32_t tcp_flags : 9;
+
+	uint32_t reserved : 20;
+	uint64_t hits;
+	uint64_t bytes;
+};
+
+/*
+ * FLOW_ACTION_TYPE_MARK specification
+ */
+struct flow_action_mark {
+	uint32_t id;	/* mark flow with this ID */
+};
+
+/*
+ * FLOW_ACTION_TYPE_TAG specification
+ */
+struct flow_action_tag {
+	uint32_t data;	/* tag flow with this value */
+	uint32_t mask;	/* bit-mask applied to "data" */
+	uint8_t index;	/* index of tag to set */
+};
+
+/*
+ * FLOW_ACTION_TYPE_PORT_ID specification
+ */
+struct flow_action_port_id {
+	uint32_t __rte_flags;	/* not used but to be binary compatible with rte flow */
+	uint32_t id;
+};
+
+/*
+ * FLOW_ACTION_TYPE_QUEUE
+ */
+struct flow_action_queue {
+	uint16_t index;
+};
+
+/*
+ * FLOW_ACTION_TYPE_JUMP
+ */
+struct flow_action_jump {
+	uint32_t group;
+};
+
+/*
+ * FLOW_ACTION_TYPE_METER
+ */
+struct flow_action_meter {
+	uint32_t mtr_id;
+};
+
+/*
+ * FLOW_ACTION_TYPE_RAW_ENCAP
+ */
+struct flow_action_raw_encap {
+	uint8_t *data;
+	uint8_t *preserve;
+	size_t size;
+	struct flow_elem items[RAW_ENCAP_DECAP_ELEMS_MAX];
+	int item_count;
+};
+
+/*
+ * FLOW_ACTION_TYPE_RAW_DECAP
+ */
+struct flow_action_raw_decap {
+	uint8_t *data;
+	size_t size;
+	struct flow_elem items[RAW_ENCAP_DECAP_ELEMS_MAX];
+	int item_count;
+};
+
+/*
+ * Field IDs for MODIFY_FIELD action.
+ */
+enum flow_field_id {
+	FLOW_FIELD_START = 0,	/* Start of a packet. */
+	FLOW_FIELD_MAC_DST,	/* Destination MAC Address. */
+	FLOW_FIELD_MAC_SRC,	/* Source MAC Address. */
+	FLOW_FIELD_VLAN_TYPE,	/* 802.1Q Tag Identifier. */
+	FLOW_FIELD_VLAN_ID,	/* 802.1Q VLAN Identifier. */
+	FLOW_FIELD_MAC_TYPE,	/* EtherType. */
+	FLOW_FIELD_IPV4_DSCP,	/* IPv4 DSCP. */
+	FLOW_FIELD_IPV4_TTL,	/* IPv4 Time To Live. */
+	FLOW_FIELD_IPV4_SRC,	/* IPv4 Source Address. */
+	FLOW_FIELD_IPV4_DST,	/* IPv4 Destination Address. */
+	FLOW_FIELD_IPV6_DSCP,	/* IPv6 DSCP. */
+	FLOW_FIELD_IPV6_HOPLIMIT,	/* IPv6 Hop Limit. */
+	FLOW_FIELD_IPV6_SRC,	/* IPv6 Source Address. */
+	FLOW_FIELD_IPV6_DST,	/* IPv6 Destination Address. */
+	FLOW_FIELD_TCP_PORT_SRC,/* TCP Source Port Number. */
+	FLOW_FIELD_TCP_PORT_DST,/* TCP Destination Port Number. */
+	FLOW_FIELD_TCP_SEQ_NUM,	/* TCP Sequence Number. */
+	FLOW_FIELD_TCP_ACK_NUM,	/* TCP Acknowledgment Number. */
+	FLOW_FIELD_TCP_FLAGS,	/* TCP Flags. */
+	FLOW_FIELD_UDP_PORT_SRC,/* UDP Source Port Number. */
+	FLOW_FIELD_UDP_PORT_DST,/* UDP Destination Port Number. */
+	FLOW_FIELD_VXLAN_VNI,	/* VXLAN Network Identifier. */
+	FLOW_FIELD_GENEVE_VNI,	/* GENEVE Network Identifier. */
+	FLOW_FIELD_GTP_TEID,	/* GTP Tunnel Endpoint Identifier. */
+	FLOW_FIELD_TAG,	/* Tag value. */
+	FLOW_FIELD_MARK,/* Mark value. */
+	FLOW_FIELD_META,/* Metadata value. */
+	FLOW_FIELD_POINTER,	/* Memory pointer. */
+	FLOW_FIELD_VALUE,	/* Immediate value. */
+	FLOW_FIELD_IPV4_ECN,	/* IPv4 ECN. */
+	FLOW_FIELD_IPV6_ECN,	/* IPv6 ECN. */
+	FLOW_FIELD_GTP_PSC_QFI,	/* GTP QFI. */
+	FLOW_FIELD_METER_COLOR,	/* Meter color marker. */
+};
+
+/*
+ * Field description for MODIFY_FIELD action.
+ */
+struct flow_action_modify_data {
+	enum flow_field_id field;	/* Field or memory type ID. */
+	union {
+		struct {
+			/* Encapsulation level or tag index. */
+			uint32_t level;
+			/* Number of bits to skip from a field. */
+			uint32_t offset;
+		};
+		/*
+		 * Immediate value for FLOW_FIELD_VALUE, presented in the
+		 * same byte order and length as in relevant rte_flow_item_xxx.
+		 */
+		uint8_t value[16];
+		/*
+		 * Memory address for FLOW_FIELD_POINTER, memory layout
+		 * should be the same as for relevant field in the
+		 * rte_flow_item_xxx structure.
+		 */
+		void *pvalue;
+	};
+};
+
+/*
+ * Operation types for MODIFY_FIELD action.
+ */
+enum flow_modify_op {
+	FLOW_MODIFY_SET = 0,
+	FLOW_MODIFY_ADD,
+	FLOW_MODIFY_SUB,
+};
+
+/*
+ * FLOW_ACTION_TYPE_MODIFY_FIELD
+ */
+struct flow_action_modify_field {
+	enum flow_modify_op operation;
+	struct flow_action_modify_data dst;
+	struct flow_action_modify_data src;
+	uint32_t width;
+};
+
+struct flow_action {
+	enum flow_action_type type;
+	const void *conf;
+};
+
+enum flow_error_e {
+	FLOW_ERROR_NONE,
+	FLOW_ERROR_SUCCESS,
+	FLOW_ERROR_GENERAL
+};
+
+struct flow_error {
+	enum flow_error_e type;
+	const char *message;
+};
+
+enum flow_lag_cmd {
+	FLOW_LAG_SET_ENTRY,
+	FLOW_LAG_SET_ALL,
+	FLOW_LAG_SET_BALANCE,
+};
+
+/*
+ * Tunnel definition for DPDK RTE tunnel helper function support
+ */
+struct tunnel_cfg_s {
+	union {
+		struct {
+			uint32_t src_ip;/* BE */
+			uint32_t dst_ip;/* BE */
+		} v4;
+		struct {
+			uint8_t src_ip[16];
+			uint8_t dst_ip[16];
+		} v6;
+		struct {
+			uint64_t src_ip[2];
+			uint64_t dst_ip[2];
+		} v6_long;
+	};
+	int ipversion;
+	uint16_t s_port;/* BE */
+	uint16_t d_port;/* BE */
+	int tun_type;
+};
+
+struct flow_eth_dev;	/* port device */
+struct flow_handle;
+
+struct flow_pattern_template;
+struct flow_actions_template;
+struct flow_template_table;
+
+/*
+ * Device Management API
+ */
+int flow_reset_nic_dev(uint8_t adapter_no);
+
+struct flow_eth_dev *flow_get_eth_dev(uint8_t adapter_no,
+	uint8_t hw_port_no,
+	uint32_t port_id,
+	int alloc_rx_queues,
+	struct flow_queue_id_s queue_ids[],
+	int *rss_target_id,
+	enum flow_eth_dev_profile flow_profile,
+	uint32_t exception_path);
+
+int flow_eth_dev_add_queue(struct flow_eth_dev *eth_dev, struct flow_queue_id_s *queue_id);
+
+int flow_delete_eth_dev(struct flow_eth_dev *eth_dev);
+
+int flow_get_tunnel_definition(struct tunnel_cfg_s *tun, uint32_t flow_stat_id, uint8_t vport);
+
+/*
+ * NT Flow API
+ */
+int flow_validate(struct flow_eth_dev *dev,
+	const struct flow_elem item[],
+	const struct flow_action action[],
+	struct flow_error *error);
+
+struct flow_handle *flow_create(struct flow_eth_dev *dev,
+	const struct flow_attr *attr,
+	const struct flow_elem item[],
+	const struct flow_action action[],
+	struct flow_error *error);
+
+int flow_destroy(struct flow_eth_dev *dev, struct flow_handle *flow, struct flow_error *error);
+
+int flow_flush(struct flow_eth_dev *dev, uint16_t caller_id, struct flow_error *error);
+
+int flow_actions_update(struct flow_eth_dev *dev,
+	struct flow_handle *flow,
+	const struct flow_action action[],
+	struct flow_error *error);
+
+int flow_query(struct flow_eth_dev *dev,
+	struct flow_handle *flow,
+	const struct flow_action *action,
+	void **data,
+	uint32_t *length,
+	struct flow_error *error);
+
+int flow_dev_dump(struct flow_eth_dev *dev,
+	struct flow_handle *flow,
+	uint16_t caller_id,
+	FILE *file,
+	struct flow_error *error);
+
+int flow_get_aged_flows(struct flow_eth_dev *dev,
+	void **context,
+	uint32_t nb_contexts,
+	struct flow_error *error);
+
+/*
+ * NT Flow asynchronous operations API
+ */
+struct flow_port_info {
+	/* maximum number of queues for asynchronous operations. */
+	uint32_t max_nb_queues;
+	/* maximum number of counters. see RTE_FLOW_ACTION_TYPE_COUNT */
+	uint32_t max_nb_counters;
+	/* maximum number of aging objects. see RTE_FLOW_ACTION_TYPE_AGE */
+	uint32_t max_nb_aging_objects;
+	/* maximum number traffic meters. see RTE_FLOW_ACTION_TYPE_METER */
+	uint32_t max_nb_meters;
+	/* maximum number connection trackings. see RTE_FLOW_ACTION_TYPE_CONNTRACK */
+	uint32_t max_nb_conn_tracks;
+	uint32_t supported_flags;	/* port supported flags (RTE_FLOW_PORT_FLAG_*). */
+};
+
+struct flow_queue_info {
+	uint32_t max_size;	/* maximum number of operations a queue can hold. */
+};
+
+struct flow_op_attr {
+	/* when set, the requested action will not be sent to the HW immediately. */
+	uint32_t postpone : 1;
+};
+
+struct flow_port_attr {
+	/* number of counters to configure. see RTE_FLOW_ACTION_TYPE_COUNT */
+	uint32_t nb_counters;
+	/* number of aging objects to configure. see RTE_FLOW_ACTION_TYPE_AGE */
+	uint32_t nb_aging_objects;
+	/* number of traffic meters to configure. see RTE_FLOW_ACTION_TYPE_METER */
+	uint32_t nb_meters;
+	/* number of connection trackings to configure. see RTE_FLOW_ACTION_TYPE_CONNTRACK */
+	uint32_t nb_conn_tracks;
+	uint32_t flags;	/* Port flags (RTE_FLOW_PORT_FLAG_*). */
+};
+
+struct flow_queue_attr {
+	uint32_t size;	/* number of flow rule operations a queue can hold. */
+};
+
+struct flow_pattern_template_attr {
+	/**
+	 * Relaxed matching policy.
+	 * - If 1, matching is performed only on items with the mask member set
+	 * and matching on protocol layers specified without any masks is skipped.
+	 * - If 0, matching on protocol layers specified without any masks is done
+	 * as well. This is the standard behaviour of Flow API now.
+	 */
+	uint32_t relaxed_matching : 1;
+	/* Flow direction for the pattern template. At least one direction must be specified. */
+	uint32_t ingress : 1;	/* pattern valid for rules applied to ingress traffic. */
+	uint32_t egress : 1;	/* pattern valid for rules applied to egress traffic. */
+	uint32_t transfer : 1;	/* pattern valid for rules applied to transfer traffic. */
+	uint32_t reserved : 28;
+	uint16_t caller_id;	/* Unique ID of caller application. */
+};
+
+struct flow_actions_template_attr {
+	/* Flow direction for the actions template. At least one direction must be specified. */
+	uint32_t ingress : 1;	/* action valid for rules applied to ingress traffic. */
+	uint32_t egress : 1;	/* action valid for rules applied to egress traffic. */
+	uint32_t transfer : 1;	/* action valid for rules applied to transfer traffic. */
+	uint32_t reserved : 29;
+	uint16_t caller_id;	/* Unique ID of caller application. */
+};
+
+struct async_flow_attr {
+	/* priority group. */
+	uint32_t group;
+	/* rule priority level within group. */
+	uint32_t priority;
+	/* the rule in question applies to ingress traffic (non-"transfer"). */
+	uint32_t ingress : 1;
+	/* the rule in question applies to egress traffic (non-"transfer"). */
+	uint32_t egress : 1;
+	/*
+	 * managing "transfer" flows requires that the user
+	 * communicate them through a suitable port.
+	 */
+	uint32_t transfer : 1;
+	uint32_t reserved : 29;	/* reserved, must be zero. */
+};
+
+struct flow_template_table_attr {
+	/* flow attributes to be used in each rule generated from this table. */
+	struct async_flow_attr flow_attr;
+	uint32_t nb_flows;	/* maximum number of flow rules that this table holds. */
+	uint16_t forced_vlan_vid;	/* Forced VLAN VID that filter must match. Ignored if 0. */
+	uint16_t caller_id;	/* Unique ID of caller application. */
+};
+
+enum flow_op_status {
+	FLOW_OP_SUCCESS,/* the operation was completed successfully. */
+	FLOW_OP_ERROR,	/* the operation was not completed successfully. */
+};
+
+struct flow_op_result {
+	/* returns the status of the operation that this completion signals. */
+	enum flow_op_status status;
+	void *user_data;/* the user data that will be returned on the completion events. */
+};
+
+struct flow_indir_action_conf {
+	uint32_t ingress : 1;	/* action valid for rules applied to ingress traffic. */
+	uint32_t egress : 1;	/* action valid for rules applied to egress traffic. */
+	/* action is valid for ransfer traffic; otherwise, for non-transfer traffic. */
+	uint32_t transfer : 1;
+};
+
+int flow_info_get(struct flow_eth_dev *dev, struct flow_port_info *port_info,
+	struct flow_queue_info *queue_info, struct flow_error *error);
+
+int flow_configure(struct flow_eth_dev *dev, uint8_t caller_id,
+	const struct flow_port_attr *port_attr, uint16_t nb_queue,
+	const struct flow_queue_attr *queue_attr[], struct flow_error *error);
+
+struct flow_pattern_template *
+flow_pattern_template_create(struct flow_eth_dev *dev,
+	const struct flow_pattern_template_attr *template_attr,
+	const struct flow_elem pattern[], struct flow_error *error);
+
+int flow_pattern_template_destroy(struct flow_eth_dev *dev,
+	struct flow_pattern_template *pattern_template,
+	struct flow_error *error);
+
+struct flow_actions_template *
+flow_actions_template_create(struct flow_eth_dev *dev,
+	const struct flow_actions_template_attr *template_attr,
+	const struct flow_action actions[], const struct flow_action masks[],
+	struct flow_error *error);
+
+int flow_actions_template_destroy(struct flow_eth_dev *dev,
+	struct flow_actions_template *actions_template,
+	struct flow_error *error);
+
+struct flow_template_table *flow_template_table_create(struct flow_eth_dev *dev,
+	const struct flow_template_table_attr *table_attr,
+	struct flow_pattern_template *pattern_templates[], uint8_t nb_pattern_templates,
+	struct flow_actions_template *actions_templates[], uint8_t nb_actions_templates,
+	struct flow_error *error);
+
+int flow_template_table_destroy(struct flow_eth_dev *dev,
+	struct flow_template_table *template_table,
+	struct flow_error *error);
+
+struct flow_handle *
+flow_async_create(struct flow_eth_dev *dev, uint32_t queue_id, const struct flow_op_attr *op_attr,
+	struct flow_template_table *template_table, const struct flow_elem pattern[],
+	uint8_t pattern_template_index, const struct flow_action actions[],
+	uint8_t actions_template_index, void *user_data, struct flow_error *error);
+
+int flow_async_destroy(struct flow_eth_dev *dev, uint32_t queue_id,
+	const struct flow_op_attr *op_attr, struct flow_handle *flow,
+	void *user_data, struct flow_error *error);
+
+int flow_push(struct flow_eth_dev *dev, uint32_t queue_id, struct flow_error *error);
+
+int flow_pull(struct flow_eth_dev *dev, uint32_t queue_id, struct flow_op_result res[],
+	uint16_t n_res, struct flow_error *error);
+
+/*
+ * NT Flow FLM Meter API
+ */
+int flow_mtr_supported(struct flow_eth_dev *dev);
+
+uint64_t flow_mtr_meter_policy_n_max(void);
+
+int flow_mtr_set_profile(struct flow_eth_dev *dev, uint32_t profile_id, uint64_t bucket_rate_a,
+	uint64_t bucket_size_a, uint64_t bucket_rate_b, uint64_t bucket_size_b);
+
+int flow_mtr_set_policy(struct flow_eth_dev *dev, uint32_t policy_id, int drop);
+
+int flow_mtr_create_meter(struct flow_eth_dev *dev, uint8_t caller_id, uint32_t mtr_id,
+	uint32_t profile_id, uint32_t policy_id, uint64_t stats_mask);
+
+int flow_mtr_probe_meter(struct flow_eth_dev *dev, uint8_t caller_id, uint32_t mtr_id);
+
+int flow_mtr_destroy_meter(struct flow_eth_dev *dev, uint8_t caller_id, uint32_t mtr_id);
+
+int flm_mtr_adjust_stats(struct flow_eth_dev *dev, uint8_t caller_id, uint32_t mtr_id,
+	uint32_t adjust_value);
+
+uint32_t flow_mtr_meters_supported(struct flow_eth_dev *dev, uint8_t caller_id);
+
+void flm_setup_queues(void);
+void flm_free_queues(void);
+uint32_t flm_lrn_update(struct flow_eth_dev *dev, uint32_t *inf_cnt);
+
+uint32_t flm_mtr_update_stats(struct flow_eth_dev *dev, uint32_t *inf_cnt);
+void flm_mtr_read_stats(struct flow_eth_dev *dev,
+	uint8_t caller_id,
+	uint32_t id,
+	uint64_t *stats_mask,
+	uint64_t *green_pkt,
+	uint64_t *green_bytes,
+	int clear);
+
+uint32_t flm_update(struct flow_eth_dev *dev);
+
+/*
+ * Config API
+ */
+int flow_set_mtu_inline(struct flow_eth_dev *dev, uint32_t port, uint16_t mtu);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif	/* _STREAM_BINARY_FLOW_API_H_ */
-- 
2.44.0