[dpdk-dev] [PATCH v9 1/4] librte_flow_classify: add flow classify library

Bernard Iremonger bernard.iremonger at intel.com
Sun Oct 22 15:32:01 CEST 2017


From: Ferruh Yigit <ferruh.yigit at intel.com>

The following APIs's are implemented in the
librte_flow_classify library:

rte_flow_classifier_create
rte_flow_classifier_free
rte_flow_classifier_query
rte_flow_classify_table_create
rte_flow_classify_table_entry_add
rte_flow_classify_table_entry_delete

The following librte_table API's are used:
f_create to create a table.
f_add to add a rule to the table.
f_del to delete a rule from the table.
f_free to free a table
f_lookup to match packets with the rules.

The library supports counting of IPv4 five tupple packets only,
ie IPv4 UDP, TCP and SCTP packets.

Signed-off-by: Ferruh Yigit <ferruh.yigit at intel.com>
Signed-off-by: Bernard Iremonger <bernard.iremonger at intel.com>
---
 MAINTAINERS                                        |   7 +
 config/common_base                                 |   6 +
 doc/api/doxy-api-index.md                          |   1 +
 doc/api/doxy-api.conf                              |   1 +
 lib/Makefile                                       |   3 +
 lib/librte_eal/common/include/rte_log.h            |   1 +
 lib/librte_flow_classify/Makefile                  |  51 ++
 lib/librte_flow_classify/rte_flow_classify.c       | 685 +++++++++++++++++++++
 lib/librte_flow_classify/rte_flow_classify.h       | 285 +++++++++
 lib/librte_flow_classify/rte_flow_classify_parse.c | 546 ++++++++++++++++
 lib/librte_flow_classify/rte_flow_classify_parse.h |  74 +++
 .../rte_flow_classify_version.map                  |  12 +
 mk/rte.app.mk                                      |   1 +
 13 files changed, 1673 insertions(+)
 create mode 100644 lib/librte_flow_classify/Makefile
 create mode 100644 lib/librte_flow_classify/rte_flow_classify.c
 create mode 100644 lib/librte_flow_classify/rte_flow_classify.h
 create mode 100644 lib/librte_flow_classify/rte_flow_classify_parse.c
 create mode 100644 lib/librte_flow_classify/rte_flow_classify_parse.h
 create mode 100644 lib/librte_flow_classify/rte_flow_classify_version.map

diff --git a/MAINTAINERS b/MAINTAINERS
index 2a58378..0981793 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -735,6 +735,13 @@ F: doc/guides/prog_guide/pdump_lib.rst
 F: app/pdump/
 F: doc/guides/tools/pdump.rst
 
+Flow classify
+M: Bernard Iremonger <bernard.iremonger at intel.com>
+F: lib/librte_flow_classify/
+F: test/test/test_flow_classify*
+F: examples/flow_classify/
+F: doc/guides/sample_app_ug/flow_classify.rst
+F: doc/guides/prog_guide/flow_classify_lib.rst
 
 Packet Framework
 ----------------
diff --git a/config/common_base b/config/common_base
index d9471e8..e1079aa 100644
--- a/config/common_base
+++ b/config/common_base
@@ -707,6 +707,12 @@ CONFIG_RTE_LIBRTE_GSO=y
 CONFIG_RTE_LIBRTE_METER=y
 
 #
+# Compile librte_classify
+#
+CONFIG_RTE_LIBRTE_FLOW_CLASSIFY=y
+CONFIG_RTE_LIBRTE_CLASSIFY_DEBUG=n
+
+#
 # Compile librte_sched
 #
 CONFIG_RTE_LIBRTE_SCHED=y
diff --git a/doc/api/doxy-api-index.md b/doc/api/doxy-api-index.md
index 990815f..e4468d0 100644
--- a/doc/api/doxy-api-index.md
+++ b/doc/api/doxy-api-index.md
@@ -111,6 +111,7 @@ The public API headers are grouped by topics:
   [ACL]                (@ref rte_acl.h),
   [EFD]                (@ref rte_efd.h),
   [member]             (@ref rte_member.h)
+  [flow_classify]      (@ref rte_flow_classify.h),
 
 - **QoS**:
   [metering]           (@ref rte_meter.h),
diff --git a/doc/api/doxy-api.conf b/doc/api/doxy-api.conf
index 9e9fa56..9edb6fd 100644
--- a/doc/api/doxy-api.conf
+++ b/doc/api/doxy-api.conf
@@ -48,6 +48,7 @@ INPUT                   = doc/api/doxy-api-index.md \
                           lib/librte_efd \
                           lib/librte_ether \
                           lib/librte_eventdev \
+                          lib/librte_flow_classify \
                           lib/librte_gro \
                           lib/librte_gso \
                           lib/librte_hash \
diff --git a/lib/Makefile b/lib/Makefile
index 86d475f..aba2593 100644
--- a/lib/Makefile
+++ b/lib/Makefile
@@ -83,6 +83,9 @@ DIRS-$(CONFIG_RTE_LIBRTE_POWER) += librte_power
 DEPDIRS-librte_power := librte_eal
 DIRS-$(CONFIG_RTE_LIBRTE_METER) += librte_meter
 DEPDIRS-librte_meter := librte_eal
+DIRS-$(CONFIG_RTE_LIBRTE_FLOW_CLASSIFY) += librte_flow_classify
+DEPDIRS-librte_flow_classify := librte_eal librte_ether librte_net
+DEPDIRS-librte_flow_classify += librte_table librte_acl librte_port
 DIRS-$(CONFIG_RTE_LIBRTE_SCHED) += librte_sched
 DEPDIRS-librte_sched := librte_eal librte_mempool librte_mbuf librte_net
 DEPDIRS-librte_sched += librte_timer
diff --git a/lib/librte_eal/common/include/rte_log.h b/lib/librte_eal/common/include/rte_log.h
index 2fa1199..67209ae 100644
--- a/lib/librte_eal/common/include/rte_log.h
+++ b/lib/librte_eal/common/include/rte_log.h
@@ -88,6 +88,7 @@ struct rte_logs {
 #define RTE_LOGTYPE_EFD       18 /**< Log related to EFD. */
 #define RTE_LOGTYPE_EVENTDEV  19 /**< Log related to eventdev. */
 #define RTE_LOGTYPE_GSO       20 /**< Log related to GSO. */
+#define RTE_LOGTYPE_CLASSIFY  21 /**< Log related to flow classify. */
 
 /* these log types can be used in an application */
 #define RTE_LOGTYPE_USER1     24 /**< User-defined log type 1. */
diff --git a/lib/librte_flow_classify/Makefile b/lib/librte_flow_classify/Makefile
new file mode 100644
index 0000000..7863a0c
--- /dev/null
+++ b/lib/librte_flow_classify/Makefile
@@ -0,0 +1,51 @@
+#   BSD LICENSE
+#
+#   Copyright(c) 2017 Intel Corporation. All rights reserved.
+#   All rights reserved.
+#
+#   Redistribution and use in source and binary forms, with or without
+#   modification, are permitted provided that the following conditions
+#   are met:
+#
+#     * Redistributions of source code must retain the above copyright
+#       notice, this list of conditions and the following disclaimer.
+#     * Redistributions in binary form must reproduce the above copyright
+#       notice, this list of conditions and the following disclaimer in
+#       the documentation and/or other materials provided with the
+#       distribution.
+#     * Neither the name of Intel Corporation nor the names of its
+#       contributors may be used to endorse or promote products derived
+#       from this software without specific prior written permission.
+#
+#   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+#   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+#   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+#   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+#   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+#   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+#   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+#   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+#   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+#   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+#   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+
+include $(RTE_SDK)/mk/rte.vars.mk
+
+# library name
+LIB = librte_flow_classify.a
+
+CFLAGS += -O3
+CFLAGS += $(WERROR_FLAGS) -I$(SRCDIR)
+
+EXPORT_MAP := rte_flow_classify_version.map
+
+LIBABIVER := 1
+
+# all source are stored in SRCS-y
+SRCS-$(CONFIG_RTE_LIBRTE_FLOW_CLASSIFY) += rte_flow_classify.c
+SRCS-$(CONFIG_RTE_LIBRTE_FLOW_CLASSIFY) += rte_flow_classify_parse.c
+
+# install this header file
+SYMLINK-$(CONFIG_RTE_LIBRTE_FLOW_CLASSIFY)-include := rte_flow_classify.h
+
+include $(RTE_SDK)/mk/rte.lib.mk
diff --git a/lib/librte_flow_classify/rte_flow_classify.c b/lib/librte_flow_classify/rte_flow_classify.c
new file mode 100644
index 0000000..794b61e
--- /dev/null
+++ b/lib/librte_flow_classify/rte_flow_classify.c
@@ -0,0 +1,685 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright(c) 2017 Intel Corporation. All rights reserved.
+ *   All rights reserved.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in
+ *       the documentation and/or other materials provided with the
+ *       distribution.
+ *     * Neither the name of Intel Corporation nor the names of its
+ *       contributors may be used to endorse or promote products derived
+ *       from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <rte_flow_classify.h>
+#include "rte_flow_classify_parse.h"
+#include <rte_flow_driver.h>
+#include <rte_table_acl.h>
+#include <stdbool.h>
+
+static struct rte_eth_ntuple_filter ntuple_filter;
+static uint32_t unique_id = 1;
+
+
+struct rte_flow_classify_table_entry {
+	/* meta-data for classify rule */
+	uint32_t rule_id;
+
+	/* Start of table entry area for user defined meta data */
+	__extension__ uint8_t meta_data[0];
+};
+
+struct rte_table {
+	/* Input parameters */
+	struct rte_table_ops ops;
+	uint32_t entry_size;
+	enum rte_flow_classify_table_type type;
+
+	/* Handle to the low-level table object */
+	void *h_table;
+};
+
+#define RTE_FLOW_CLASSIFIER_MAX_NAME_SZ 256
+
+struct rte_flow_classifier {
+	/* Input parameters */
+	char name[RTE_FLOW_CLASSIFIER_MAX_NAME_SZ];
+	int socket_id;
+	enum rte_flow_classify_table_type type;
+
+	/* Internal tables */
+	struct rte_table tables[RTE_FLOW_CLASSIFY_TABLE_MAX];
+	uint32_t num_tables;
+	uint16_t nb_pkts;
+	struct rte_flow_classify_table_entry
+		*entries[RTE_PORT_IN_BURST_SIZE_MAX];
+} __rte_cache_aligned;
+
+enum {
+	PROTO_FIELD_IPV4,
+	SRC_FIELD_IPV4,
+	DST_FIELD_IPV4,
+	SRCP_FIELD_IPV4,
+	DSTP_FIELD_IPV4,
+	NUM_FIELDS_IPV4
+};
+
+struct acl_keys {
+	struct rte_table_acl_rule_add_params key_add; /* add key */
+	struct rte_table_acl_rule_delete_params	key_del; /* delete key */
+};
+
+struct classify_rules {
+	enum rte_flow_classify_rule_type type;
+	union {
+		struct rte_flow_classify_ipv4_5tuple ipv4_5tuple;
+	} u;
+};
+
+struct rte_flow_classify_rule {
+	uint32_t id; /* unique ID of classify rule */
+	struct rte_flow_action action; /* action when match found */
+	struct classify_rules rules; /* union of rules */
+	union {
+		struct acl_keys key;
+	} u;
+	int key_found;   /* rule key found in table */
+	void *entry;     /* pointer to buffer to hold rule meta data */
+	void *entry_ptr; /* handle to the table entry for rule meta data */
+};
+
+static int
+rte_flow_classify_parse_flow(
+		   const struct rte_flow_attr *attr,
+		   const struct rte_flow_item pattern[],
+		   const struct rte_flow_action actions[],
+		   struct rte_flow_error *error)
+{
+	struct rte_flow_item *items;
+	parse_filter_t parse_filter;
+	uint32_t item_num = 0;
+	uint32_t i = 0;
+	int ret;
+
+	memset(&ntuple_filter, 0, sizeof(ntuple_filter));
+
+	/* Get the non-void item number of pattern */
+	while ((pattern + i)->type != RTE_FLOW_ITEM_TYPE_END) {
+		if ((pattern + i)->type != RTE_FLOW_ITEM_TYPE_VOID)
+			item_num++;
+		i++;
+	}
+	item_num++;
+
+	items = malloc(item_num * sizeof(struct rte_flow_item));
+	if (!items) {
+		rte_flow_error_set(error, ENOMEM,
+				RTE_FLOW_ERROR_TYPE_ITEM_NUM,
+				NULL, "No memory for pattern items.");
+		return -ENOMEM;
+	}
+
+	memset(items, 0, item_num * sizeof(struct rte_flow_item));
+	classify_pattern_skip_void_item(items, pattern);
+
+	parse_filter = classify_find_parse_filter_func(items);
+	if (!parse_filter) {
+		rte_flow_error_set(error, EINVAL,
+				RTE_FLOW_ERROR_TYPE_ITEM,
+				pattern, "Unsupported pattern");
+		free(items);
+		return -EINVAL;
+	}
+
+	ret = parse_filter(attr, items, actions, &ntuple_filter, error);
+	free(items);
+	return ret;
+}
+
+#ifdef RTE_LIBRTE_CLASSIFY_DEBUG
+#define uint32_t_to_char(ip, a, b, c, d) do {\
+		*a = (unsigned char)(ip >> 24 & 0xff);\
+		*b = (unsigned char)(ip >> 16 & 0xff);\
+		*c = (unsigned char)(ip >> 8 & 0xff);\
+		*d = (unsigned char)(ip & 0xff);\
+	} while (0)
+
+static inline void
+print_acl_ipv4_key_add(struct rte_table_acl_rule_add_params *key)
+{
+	unsigned char a, b, c, d;
+
+	printf("%s:    0x%02hhx/0x%hhx ", __func__,
+		key->field_value[PROTO_FIELD_IPV4].value.u8,
+		key->field_value[PROTO_FIELD_IPV4].mask_range.u8);
+
+	uint32_t_to_char(key->field_value[SRC_FIELD_IPV4].value.u32,
+			&a, &b, &c, &d);
+	printf(" %hhu.%hhu.%hhu.%hhu/0x%x ", a, b, c, d,
+			key->field_value[SRC_FIELD_IPV4].mask_range.u32);
+
+	uint32_t_to_char(key->field_value[DST_FIELD_IPV4].value.u32,
+			&a, &b, &c, &d);
+	printf("%hhu.%hhu.%hhu.%hhu/0x%x ", a, b, c, d,
+			key->field_value[DST_FIELD_IPV4].mask_range.u32);
+
+	printf("%hu : 0x%x %hu : 0x%x",
+		key->field_value[SRCP_FIELD_IPV4].value.u16,
+		key->field_value[SRCP_FIELD_IPV4].mask_range.u16,
+		key->field_value[DSTP_FIELD_IPV4].value.u16,
+		key->field_value[DSTP_FIELD_IPV4].mask_range.u16);
+
+	printf(" priority: 0x%x\n", key->priority);
+}
+
+static inline void
+print_acl_ipv4_key_delete(struct rte_table_acl_rule_delete_params *key)
+{
+	unsigned char a, b, c, d;
+
+	printf("%s: 0x%02hhx/0x%hhx ", __func__,
+		key->field_value[PROTO_FIELD_IPV4].value.u8,
+		key->field_value[PROTO_FIELD_IPV4].mask_range.u8);
+
+	uint32_t_to_char(key->field_value[SRC_FIELD_IPV4].value.u32,
+			&a, &b, &c, &d);
+	printf(" %hhu.%hhu.%hhu.%hhu/0x%x ", a, b, c, d,
+			key->field_value[SRC_FIELD_IPV4].mask_range.u32);
+
+	uint32_t_to_char(key->field_value[DST_FIELD_IPV4].value.u32,
+			&a, &b, &c, &d);
+	printf("%hhu.%hhu.%hhu.%hhu/0x%x ", a, b, c, d,
+			key->field_value[DST_FIELD_IPV4].mask_range.u32);
+
+	printf("%hu : 0x%x %hu : 0x%x\n",
+		key->field_value[SRCP_FIELD_IPV4].value.u16,
+		key->field_value[SRCP_FIELD_IPV4].mask_range.u16,
+		key->field_value[DSTP_FIELD_IPV4].value.u16,
+		key->field_value[DSTP_FIELD_IPV4].mask_range.u16);
+}
+#endif
+
+static int
+rte_flow_classifier_check_params(struct rte_flow_classifier_params *params)
+{
+	if (params == NULL) {
+		RTE_LOG(ERR, CLASSIFY,
+			"%s: Incorrect value for parameter params\n", __func__);
+		return -EINVAL;
+	}
+
+	/* name */
+	if (params->name == NULL) {
+		RTE_LOG(ERR, CLASSIFY,
+			"%s: Incorrect value for parameter name\n", __func__);
+		return -EINVAL;
+	}
+
+	/* socket */
+	if ((params->socket_id < 0) ||
+	    (params->socket_id >= RTE_MAX_NUMA_NODES)) {
+		RTE_LOG(ERR, CLASSIFY,
+			"%s: Incorrect value for parameter socket_id\n",
+			__func__);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+struct rte_flow_classifier *
+rte_flow_classifier_create(struct rte_flow_classifier_params *params)
+{
+	struct rte_flow_classifier *cls;
+	int ret;
+
+	/* Check input parameters */
+	ret = rte_flow_classifier_check_params(params);
+	if (ret != 0) {
+		RTE_LOG(ERR, CLASSIFY,
+			"%s: flow classifier params check failed (%d)\n",
+			__func__, ret);
+		return NULL;
+	}
+
+	/* Allocate memory for the flow classifier */
+	cls = rte_zmalloc_socket("FLOW_CLASSIFIER",
+			sizeof(struct rte_flow_classifier),
+			RTE_CACHE_LINE_SIZE, params->socket_id);
+
+	if (cls == NULL) {
+		RTE_LOG(ERR, CLASSIFY,
+			"%s: flow classifier memory allocation failed\n",
+			__func__);
+		return NULL;
+	}
+
+	/* Save input parameters */
+	snprintf(cls->name, RTE_FLOW_CLASSIFIER_MAX_NAME_SZ, "%s",
+			params->name);
+	cls->socket_id = params->socket_id;
+	cls->type = params->type;
+
+	/* Initialize flow classifier internal data structure */
+	cls->num_tables = 0;
+
+	return cls;
+}
+
+static void
+rte_flow_classify_table_free(struct rte_table *table)
+{
+	if (table->ops.f_free != NULL)
+		table->ops.f_free(table->h_table);
+}
+
+int
+rte_flow_classifier_free(struct rte_flow_classifier *cls)
+{
+	uint32_t i;
+
+	/* Check input parameters */
+	if (cls == NULL) {
+		RTE_LOG(ERR, CLASSIFY,
+			"%s: rte_flow_classifier parameter is NULL\n",
+			__func__);
+		return -EINVAL;
+	}
+
+	/* Free tables */
+	for (i = 0; i < cls->num_tables; i++) {
+		struct rte_table *table = &cls->tables[i];
+
+		rte_flow_classify_table_free(table);
+	}
+
+	/* Free flow classifier memory */
+	rte_free(cls);
+
+	return 0;
+}
+
+static int
+rte_table_check_params(struct rte_flow_classifier *cls,
+		struct rte_flow_classify_table_params *params,
+		uint32_t *table_id)
+{
+	if (cls == NULL) {
+		RTE_LOG(ERR, CLASSIFY,
+			"%s: flow classifier parameter is NULL\n",
+			__func__);
+		return -EINVAL;
+	}
+	if (params == NULL) {
+		RTE_LOG(ERR, CLASSIFY, "%s: params parameter is NULL\n",
+			__func__);
+		return -EINVAL;
+	}
+	if (table_id == NULL) {
+		RTE_LOG(ERR, CLASSIFY, "%s: table_id parameter is NULL\n",
+			__func__);
+		return -EINVAL;
+	}
+
+	/* ops */
+	if (params->ops == NULL) {
+		RTE_LOG(ERR, CLASSIFY, "%s: params->ops is NULL\n",
+			__func__);
+		return -EINVAL;
+	}
+
+	if (params->ops->f_create == NULL) {
+		RTE_LOG(ERR, CLASSIFY,
+			"%s: f_create function pointer is NULL\n", __func__);
+		return -EINVAL;
+	}
+
+	if (params->ops->f_lookup == NULL) {
+		RTE_LOG(ERR, CLASSIFY,
+			"%s: f_lookup function pointer is NULL\n", __func__);
+		return -EINVAL;
+	}
+
+	/* De we have room for one more table? */
+	if (cls->num_tables == RTE_FLOW_CLASSIFY_TABLE_MAX) {
+		RTE_LOG(ERR, CLASSIFY,
+			"%s: Incorrect value for num_tables parameter\n",
+			__func__);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+int
+rte_flow_classify_table_create(struct rte_flow_classifier *cls,
+	struct rte_flow_classify_table_params *params,
+	uint32_t *table_id)
+{
+	struct rte_table *table;
+	void *h_table;
+	uint32_t entry_size, id;
+	int ret;
+
+	/* Check input arguments */
+	ret = rte_table_check_params(cls, params, table_id);
+	if (ret != 0)
+		return ret;
+
+	id = cls->num_tables;
+	table = &cls->tables[id];
+
+	/* calculate table entry size */
+	entry_size = sizeof(struct rte_flow_classify_table_entry) +
+			params->table_metadata_size;
+
+	/* Create the table */
+	h_table = params->ops->f_create(params->arg_create, cls->socket_id,
+		entry_size);
+	if (h_table == NULL) {
+		RTE_LOG(ERR, CLASSIFY, "%s: Table creation failed\n", __func__);
+		return -EINVAL;
+	}
+
+	/* Commit current table to the classifier */
+	cls->num_tables++;
+	*table_id = id;
+
+	/* Save input parameters */
+	memcpy(&table->ops, params->ops, sizeof(struct rte_table_ops));
+
+	/* Initialize table internal data structure */
+	table->entry_size = entry_size;
+	table->h_table = h_table;
+
+	return 0;
+}
+
+static struct rte_flow_classify_rule *
+allocate_acl_ipv4_5tuple_rule(void)
+{
+	struct rte_flow_classify_rule *rule;
+
+	rule = malloc(sizeof(struct rte_flow_classify_rule));
+	if (!rule)
+		return rule;
+
+	memset(rule, 0, sizeof(struct rte_flow_classify_rule));
+	rule->id = unique_id++;
+	rule->rules.type = RTE_FLOW_CLASSIFY_RULE_TYPE_IPV4_5TUPLE;
+
+	memcpy(&rule->action, classify_get_flow_action(),
+	       sizeof(struct rte_flow_action));
+
+	/* key add values */
+	rule->u.key.key_add.priority = ntuple_filter.priority;
+	rule->u.key.key_add.field_value[PROTO_FIELD_IPV4].mask_range.u8 =
+			ntuple_filter.proto_mask;
+	rule->u.key.key_add.field_value[PROTO_FIELD_IPV4].value.u8 =
+			ntuple_filter.proto;
+	rule->rules.u.ipv4_5tuple.proto = ntuple_filter.proto;
+	rule->rules.u.ipv4_5tuple.proto_mask = ntuple_filter.proto_mask;
+
+	rule->u.key.key_add.field_value[SRC_FIELD_IPV4].mask_range.u32 =
+			ntuple_filter.src_ip_mask;
+	rule->u.key.key_add.field_value[SRC_FIELD_IPV4].value.u32 =
+			ntuple_filter.src_ip;
+	rule->rules.u.ipv4_5tuple.src_ip_mask = ntuple_filter.src_ip_mask;
+	rule->rules.u.ipv4_5tuple.src_ip = ntuple_filter.src_ip;
+
+	rule->u.key.key_add.field_value[DST_FIELD_IPV4].mask_range.u32 =
+			ntuple_filter.dst_ip_mask;
+	rule->u.key.key_add.field_value[DST_FIELD_IPV4].value.u32 =
+			ntuple_filter.dst_ip;
+	rule->rules.u.ipv4_5tuple.dst_ip_mask = ntuple_filter.dst_ip_mask;
+	rule->rules.u.ipv4_5tuple.dst_ip = ntuple_filter.dst_ip;
+
+	rule->u.key.key_add.field_value[SRCP_FIELD_IPV4].mask_range.u16 =
+			ntuple_filter.src_port_mask;
+	rule->u.key.key_add.field_value[SRCP_FIELD_IPV4].value.u16 =
+			ntuple_filter.src_port;
+	rule->rules.u.ipv4_5tuple.src_port_mask = ntuple_filter.src_port_mask;
+	rule->rules.u.ipv4_5tuple.src_port = ntuple_filter.src_port;
+
+	rule->u.key.key_add.field_value[DSTP_FIELD_IPV4].mask_range.u16 =
+			ntuple_filter.dst_port_mask;
+	rule->u.key.key_add.field_value[DSTP_FIELD_IPV4].value.u16 =
+			ntuple_filter.dst_port;
+	rule->rules.u.ipv4_5tuple.dst_port_mask = ntuple_filter.dst_port_mask;
+	rule->rules.u.ipv4_5tuple.dst_port = ntuple_filter.dst_port;
+
+#ifdef RTE_LIBRTE_CLASSIFY_DEBUG
+	print_acl_ipv4_key_add(&rule->u.key.key_add);
+#endif
+
+	/* key delete values */
+	memcpy(&rule->u.key.key_del.field_value[PROTO_FIELD_IPV4],
+	       &rule->u.key.key_add.field_value[PROTO_FIELD_IPV4],
+	       NUM_FIELDS_IPV4 * sizeof(struct rte_acl_field));
+
+#ifdef RTE_LIBRTE_CLASSIFY_DEBUG
+	print_acl_ipv4_key_delete(&rule->u.key.key_del);
+#endif
+	return rule;
+}
+
+struct rte_flow_classify_rule *
+rte_flow_classify_table_entry_add(struct rte_flow_classifier *cls,
+		uint32_t table_id,
+		int *key_found,
+		const struct rte_flow_attr *attr,
+		const struct rte_flow_item pattern[],
+		const struct rte_flow_action actions[],
+		struct rte_flow_error *error)
+{
+	struct rte_flow_classify_rule *rule;
+	struct rte_flow_classify_table_entry *table_entry;
+	int ret;
+
+	if (!error)
+		return NULL;
+
+	if (!cls) {
+		rte_flow_error_set(error, EINVAL,
+				RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
+				NULL, "NULL classifier.");
+		return NULL;
+	}
+
+	if (table_id >= cls->num_tables) {
+		rte_flow_error_set(error, EINVAL,
+				RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
+				NULL, "invalid table_id.");
+		return NULL;
+	}
+
+	if (key_found == NULL) {
+		rte_flow_error_set(error, EINVAL,
+				RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
+				NULL, "NULL key_found.");
+		return NULL;
+	}
+
+	if (!pattern) {
+		rte_flow_error_set(error, EINVAL,
+				RTE_FLOW_ERROR_TYPE_ITEM_NUM,
+				NULL, "NULL pattern.");
+		return NULL;
+	}
+
+	if (!actions) {
+		rte_flow_error_set(error, EINVAL,
+				RTE_FLOW_ERROR_TYPE_ACTION_NUM,
+				NULL, "NULL action.");
+		return NULL;
+	}
+
+	if (!attr) {
+		rte_flow_error_set(error, EINVAL,
+				RTE_FLOW_ERROR_TYPE_ATTR,
+				NULL, "NULL attribute.");
+		return NULL;
+	}
+
+	/* parse attr, pattern and actions */
+	ret = rte_flow_classify_parse_flow(attr, pattern, actions, error);
+	if (ret < 0)
+		return NULL;
+
+	switch (cls->type) {
+	case RTE_FLOW_CLASSIFY_TABLE_TYPE_ACL:
+		rule = allocate_acl_ipv4_5tuple_rule();
+		if (!rule)
+			return NULL;
+		break;
+	default:
+		return NULL;
+	}
+
+	rule->entry = malloc(sizeof(struct rte_flow_classify_table_entry));
+	if (!rule->entry) {
+		free(rule);
+		return NULL;
+	}
+
+	table_entry = rule->entry;
+	table_entry->rule_id = rule->id;
+
+	if (cls->tables[table_id].ops.f_add != NULL) {
+		ret = cls->tables[table_id].ops.f_add(
+			cls->tables[table_id].h_table,
+			&rule->u.key.key_add,
+			rule->entry,
+			&rule->key_found,
+			&rule->entry_ptr);
+		if (ret) {
+			free(rule->entry);
+			free(rule);
+			return NULL;
+		}
+		*key_found = rule->key_found;
+	}
+	return rule;
+}
+
+int
+rte_flow_classify_table_entry_delete(struct rte_flow_classifier *cls,
+		uint32_t table_id,
+		struct rte_flow_classify_rule *rule)
+{
+	int ret = -EINVAL;
+
+	if (!cls || !rule || table_id >= cls->num_tables)
+		return ret;
+
+	if (cls->tables[table_id].ops.f_delete != NULL)
+		ret = cls->tables[table_id].ops.f_delete(
+			cls->tables[table_id].h_table,
+			&rule->u.key.key_del,
+			&rule->key_found,
+			&rule->entry);
+
+	return ret;
+}
+
+static int
+flow_classifier_run(struct rte_flow_classifier *cls,
+		uint32_t table_id,
+		struct rte_mbuf **pkts,
+		const uint16_t nb_pkts)
+{
+	int ret = -EINVAL;
+	uint64_t pkts_mask;
+	uint64_t lookup_hit_mask;
+
+	if (!cls || !pkts || nb_pkts == 0 || table_id >= cls->num_tables)
+		return ret;
+
+	if (cls->tables[table_id].ops.f_lookup != NULL) {
+		pkts_mask = RTE_LEN2MASK(nb_pkts, uint64_t);
+		ret = cls->tables[table_id].ops.f_lookup(
+			cls->tables[table_id].h_table,
+			pkts, pkts_mask, &lookup_hit_mask,
+			(void **)cls->entries);
+
+		if (!ret && lookup_hit_mask)
+			cls->nb_pkts = nb_pkts;
+		else
+			cls->nb_pkts = 0;
+	}
+
+	return ret;
+}
+
+static int
+action_apply(struct rte_flow_classifier *cls,
+		struct rte_flow_classify_rule *rule,
+		struct rte_flow_classify_stats *stats)
+{
+	struct rte_flow_classify_ipv4_5tuple_stats *ntuple_stats;
+	uint64_t count = 0;
+	int i;
+	int ret = -ENODATA;
+
+	switch (rule->action.type) {
+	case RTE_FLOW_ACTION_TYPE_COUNT:
+		for (i = 0; i < cls->nb_pkts; i++) {
+			if (rule->id == cls->entries[i]->rule_id)
+				count++;
+		}
+		if (count) {
+			ret = 0;
+			ntuple_stats =
+				(struct rte_flow_classify_ipv4_5tuple_stats *)
+				stats->stats;
+			ntuple_stats->counter1 = count;
+			ntuple_stats->ipv4_5tuple = rule->rules.u.ipv4_5tuple;
+		}
+		break;
+	default:
+		ret = -ENOTSUP;
+		break;
+	}
+
+	return ret;
+}
+
+int
+rte_flow_classifier_query(struct rte_flow_classifier *cls,
+		uint32_t table_id,
+		struct rte_mbuf **pkts,
+		const uint16_t nb_pkts,
+		struct rte_flow_classify_rule *rule,
+		struct rte_flow_classify_stats *stats)
+{
+	int ret = -EINVAL;
+
+	if (!cls || !rule || !stats || !pkts  || nb_pkts == 0 ||
+		table_id >= cls->num_tables)
+		return ret;
+
+	ret = flow_classifier_run(cls, table_id, pkts, nb_pkts);
+	if (!ret)
+		ret = action_apply(cls, rule, stats);
+	return ret;
+}
diff --git a/lib/librte_flow_classify/rte_flow_classify.h b/lib/librte_flow_classify/rte_flow_classify.h
new file mode 100644
index 0000000..3c97986
--- /dev/null
+++ b/lib/librte_flow_classify/rte_flow_classify.h
@@ -0,0 +1,285 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright(c) 2017 Intel Corporation. All rights reserved.
+ *   All rights reserved.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in
+ *       the documentation and/or other materials provided with the
+ *       distribution.
+ *     * Neither the name of Intel Corporation nor the names of its
+ *       contributors may be used to endorse or promote products derived
+ *       from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef _RTE_FLOW_CLASSIFY_H_
+#define _RTE_FLOW_CLASSIFY_H_
+
+/**
+ * @file
+ *
+ * RTE Flow Classify Library
+ *
+ * This library provides flow record information with some measured properties.
+ *
+ * Application should define the flow and measurement criteria (action) for it.
+ *
+ * The Library doesn't maintain any flow records itself, instead flow
+ * information is returned to upper layer only for given packets.
+ *
+ * It is application's responsibility to call rte_flow_classifier_query()
+ * for a burst of packets, just after receiving them or before transmitting
+ * them.
+ * Application should provide the flow type interested in, measurement to apply
+ * to that flow in rte_flow_classify_table_entry_add() API, and should provide
+ * the rte_flow_classifier object and storage to put results in for the
+ * rte_flow_classifier_query() API.
+ *
+ *  Usage:
+ *  - application calls rte_flow_classifier_create() to create an
+ *    rte_flow_classifier object.
+ *  - application calls rte_flow_classify_table_create() to create a table
+ *    in the rte_flow_classifier object.
+ *  - application calls rte_flow_classify_table_entry_add() to add a rule to
+ *    the table in the rte_flow_classifier object.
+ *  - application calls rte_flow_classifier_query() in a polling manner,
+ *    preferably after rte_eth_rx_burst(). This will cause the library to
+ *    match packet information to flow information with some measurements.
+ *  - rte_flow_classifier object can be destroyed when it is no longer needed
+ *    with rte_flow_classifier_free()
+ */
+
+#include <rte_ethdev.h>
+#include <rte_ether.h>
+#include <rte_flow.h>
+#include <rte_acl.h>
+#include <rte_table_acl.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/** Opaque data type for flow classifier */
+struct rte_flow_classifier;
+
+/** Opaque data type for flow classify rule */
+struct rte_flow_classify_rule;
+
+/** Flow classify rule type */
+enum rte_flow_classify_rule_type {
+	/** no type */
+	RTE_FLOW_CLASSIFY_RULE_TYPE_NONE,
+	/** IPv4 5tuple type */
+	RTE_FLOW_CLASSIFY_RULE_TYPE_IPV4_5TUPLE,
+};
+
+/** Flow classify table type */
+enum rte_flow_classify_table_type {
+	/** no type */
+	RTE_FLOW_CLASSIFY_TABLE_TYPE_NONE,
+	/** ACL type */
+	RTE_FLOW_CLASSIFY_TABLE_TYPE_ACL,
+};
+
+/**
+ * Maximum number of tables allowed for any Flow Classifier instance.
+ * The value of this parameter cannot be changed.
+ */
+#define RTE_FLOW_CLASSIFY_TABLE_MAX  64
+
+/** Parameters for flow classifier creation */
+struct rte_flow_classifier_params {
+	/** flow classifier name */
+	const char *name;
+
+	/** CPU socket ID where memory for the flow classifier and its */
+	/** elements (tables) should be allocated */
+	int socket_id;
+
+	/** Table type */
+	enum rte_flow_classify_table_type type;
+};
+
+/** Parameters for table creation */
+struct rte_flow_classify_table_params {
+	/** Table operations (specific to each table type) */
+	struct rte_table_ops *ops;
+
+	/** Opaque param to be passed to the table create operation */
+	void *arg_create;
+
+	/** Memory size to be reserved per classifier object entry for */
+	/** storing meta data */
+	uint32_t table_metadata_size;
+};
+
+/** IPv4 5-tuple data */
+struct rte_flow_classify_ipv4_5tuple {
+	uint32_t dst_ip;         /**< Destination IP address in big endian. */
+	uint32_t dst_ip_mask;    /**< Mask of destination IP address. */
+	uint32_t src_ip;         /**< Source IP address in big endian. */
+	uint32_t src_ip_mask;    /**< Mask of destination IP address. */
+	uint16_t dst_port;       /**< Destination port in big endian. */
+	uint16_t dst_port_mask;  /**< Mask of destination port. */
+	uint16_t src_port;       /**< Source Port in big endian. */
+	uint16_t src_port_mask;  /**< Mask of source port. */
+	uint8_t proto;           /**< L4 protocol. */
+	uint8_t proto_mask;      /**< Mask of L4 protocol. */
+};
+
+/**
+ * Flow stats
+ *
+ * For the count action, stats can be returned by the query API.
+ *
+ * Storage for stats is provided by application.
+ */
+struct rte_flow_classify_stats {
+	void *stats;
+};
+
+struct rte_flow_classify_ipv4_5tuple_stats {
+	/** count of packets that match IPv4 5tuple pattern */
+	uint64_t counter1;
+	/** IPv4 5tuple data */
+	struct rte_flow_classify_ipv4_5tuple ipv4_5tuple;
+};
+
+/**
+ * Flow classifier create
+ *
+ * @param params
+ *   Parameters for flow classifier creation
+ * @return
+ *   Handle to flow classifier instance on success or NULL otherwise
+ */
+struct rte_flow_classifier *
+rte_flow_classifier_create(struct rte_flow_classifier_params *params);
+
+/**
+ * Flow classifier free
+ *
+ * @param cls
+ *   Handle to flow classifier instance
+ * @return
+ *   0 on success, error code otherwise
+ */
+int
+rte_flow_classifier_free(struct rte_flow_classifier *cls);
+
+/**
+ * Flow classify table create
+ *
+ * @param cls
+ *   Handle to flow classifier instance
+ * @param params
+ *   Parameters for flow_classify table creation
+ * @param table_id
+ *   Table ID. Valid only within the scope of table IDs of the current
+ *   classifier. Only returned after a successful invocation.
+ * @return
+ *   0 on success, error code otherwise
+ */
+int
+rte_flow_classify_table_create(struct rte_flow_classifier *cls,
+		struct rte_flow_classify_table_params *params,
+		uint32_t *table_id);
+
+/**
+ * Add a flow classify rule to the flow_classifer table.
+ *
+ * @param[in] cls
+ *   Flow classifier handle
+ * @param[in] table_id
+ *   id of table
+ * @param[out] key_found
+ *  returns 1 if key present already, 0 otherwise.
+ * @param[in] attr
+ *   Flow rule attributes
+ * @param[in] pattern
+ *   Pattern specification (list terminated by the END pattern item).
+ * @param[in] actions
+ *   Associated actions (list terminated by the END pattern item).
+ * @param[out] error
+ *   Perform verbose error reporting if not NULL. Structure
+ *   initialised in case of error only.
+ * @return
+ *   A valid handle in case of success, NULL otherwise.
+ */
+struct rte_flow_classify_rule *
+rte_flow_classify_table_entry_add(struct rte_flow_classifier *cls,
+		uint32_t table_id,
+		int *key_found,
+		const struct rte_flow_attr *attr,
+		const struct rte_flow_item pattern[],
+		const struct rte_flow_action actions[],
+		struct rte_flow_error *error);
+
+/**
+ * Delete a flow classify rule from the flow_classifer table.
+ *
+ * @param[in] cls
+ *   Flow classifier handle
+ * @param[in] table_id
+ *   id of table
+ * @param[in] rule
+ *   Flow classify rule
+ * @return
+ *   0 on success, error code otherwise.
+ */
+int
+rte_flow_classify_table_entry_delete(struct rte_flow_classifier *cls,
+		uint32_t table_id,
+		struct rte_flow_classify_rule *rule);
+
+/**
+ * Query flow classifier for given rule.
+ *
+ * @param[in] cls
+ *   Flow classifier handle
+ * @param[in] table_id
+ *   id of table
+ * @param[in] pkts
+ *   Pointer to packets to process
+ * @param[in] nb_pkts
+ *   Number of packets to process
+ * @param[in] rule
+ *   Flow classify rule
+ * @param[in] stats
+ *   Flow classify stats
+ *
+ * @return
+ *   0 on success, error code otherwise.
+ */
+int
+rte_flow_classifier_query(struct rte_flow_classifier *cls,
+		uint32_t table_id,
+		struct rte_mbuf **pkts,
+		const uint16_t nb_pkts,
+		struct rte_flow_classify_rule *rule,
+		struct rte_flow_classify_stats *stats);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_FLOW_CLASSIFY_H_ */
diff --git a/lib/librte_flow_classify/rte_flow_classify_parse.c b/lib/librte_flow_classify/rte_flow_classify_parse.c
new file mode 100644
index 0000000..921a852
--- /dev/null
+++ b/lib/librte_flow_classify/rte_flow_classify_parse.c
@@ -0,0 +1,546 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright(c) 2017 Intel Corporation. All rights reserved.
+ *   All rights reserved.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in
+ *       the documentation and/or other materials provided with the
+ *       distribution.
+ *     * Neither the name of Intel Corporation nor the names of its
+ *       contributors may be used to endorse or promote products derived
+ *       from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <rte_flow_classify.h>
+#include "rte_flow_classify_parse.h"
+#include <rte_flow_driver.h>
+
+struct classify_valid_pattern {
+	enum rte_flow_item_type *items;
+	parse_filter_t parse_filter;
+};
+
+static struct rte_flow_action action;
+
+/* Pattern matched ntuple filter */
+static enum rte_flow_item_type pattern_ntuple_1[] = {
+	RTE_FLOW_ITEM_TYPE_ETH,
+	RTE_FLOW_ITEM_TYPE_IPV4,
+	RTE_FLOW_ITEM_TYPE_UDP,
+	RTE_FLOW_ITEM_TYPE_END,
+};
+
+/* Pattern matched ntuple filter */
+static enum rte_flow_item_type pattern_ntuple_2[] = {
+	RTE_FLOW_ITEM_TYPE_ETH,
+	RTE_FLOW_ITEM_TYPE_IPV4,
+	RTE_FLOW_ITEM_TYPE_TCP,
+	RTE_FLOW_ITEM_TYPE_END,
+};
+
+/* Pattern matched ntuple filter */
+static enum rte_flow_item_type pattern_ntuple_3[] = {
+	RTE_FLOW_ITEM_TYPE_ETH,
+	RTE_FLOW_ITEM_TYPE_IPV4,
+	RTE_FLOW_ITEM_TYPE_SCTP,
+	RTE_FLOW_ITEM_TYPE_END,
+};
+
+static int
+classify_parse_ntuple_filter(const struct rte_flow_attr *attr,
+			 const struct rte_flow_item pattern[],
+			 const struct rte_flow_action actions[],
+			 struct rte_eth_ntuple_filter *filter,
+			 struct rte_flow_error *error);
+
+static struct classify_valid_pattern classify_supported_patterns[] = {
+	/* ntuple */
+	{ pattern_ntuple_1, classify_parse_ntuple_filter },
+	{ pattern_ntuple_2, classify_parse_ntuple_filter },
+	{ pattern_ntuple_3, classify_parse_ntuple_filter },
+};
+
+struct rte_flow_action *
+classify_get_flow_action(void)
+{
+	return &action;
+}
+
+/* Find the first VOID or non-VOID item pointer */
+const struct rte_flow_item *
+classify_find_first_item(const struct rte_flow_item *item, bool is_void)
+{
+	bool is_find;
+
+	while (item->type != RTE_FLOW_ITEM_TYPE_END) {
+		if (is_void)
+			is_find = item->type == RTE_FLOW_ITEM_TYPE_VOID;
+		else
+			is_find = item->type != RTE_FLOW_ITEM_TYPE_VOID;
+		if (is_find)
+			break;
+		item++;
+	}
+	return item;
+}
+
+/* Skip all VOID items of the pattern */
+void
+classify_pattern_skip_void_item(struct rte_flow_item *items,
+			    const struct rte_flow_item *pattern)
+{
+	uint32_t cpy_count = 0;
+	const struct rte_flow_item *pb = pattern, *pe = pattern;
+
+	for (;;) {
+		/* Find a non-void item first */
+		pb = classify_find_first_item(pb, false);
+		if (pb->type == RTE_FLOW_ITEM_TYPE_END) {
+			pe = pb;
+			break;
+		}
+
+		/* Find a void item */
+		pe = classify_find_first_item(pb + 1, true);
+
+		cpy_count = pe - pb;
+		rte_memcpy(items, pb, sizeof(struct rte_flow_item) * cpy_count);
+
+		items += cpy_count;
+
+		if (pe->type == RTE_FLOW_ITEM_TYPE_END) {
+			pb = pe;
+			break;
+		}
+
+		pb = pe + 1;
+	}
+	/* Copy the END item. */
+	rte_memcpy(items, pe, sizeof(struct rte_flow_item));
+}
+
+/* Check if the pattern matches a supported item type array */
+static bool
+classify_match_pattern(enum rte_flow_item_type *item_array,
+		   struct rte_flow_item *pattern)
+{
+	struct rte_flow_item *item = pattern;
+
+	while ((*item_array == item->type) &&
+	       (*item_array != RTE_FLOW_ITEM_TYPE_END)) {
+		item_array++;
+		item++;
+	}
+
+	return (*item_array == RTE_FLOW_ITEM_TYPE_END &&
+		item->type == RTE_FLOW_ITEM_TYPE_END);
+}
+
+/* Find if there's parse filter function matched */
+parse_filter_t
+classify_find_parse_filter_func(struct rte_flow_item *pattern)
+{
+	parse_filter_t parse_filter = NULL;
+	uint8_t i = 0;
+
+	for (; i < RTE_DIM(classify_supported_patterns); i++) {
+		if (classify_match_pattern(classify_supported_patterns[i].items,
+					pattern)) {
+			parse_filter =
+				classify_supported_patterns[i].parse_filter;
+			break;
+		}
+	}
+
+	return parse_filter;
+}
+
+#define FLOW_RULE_MIN_PRIORITY 8
+#define FLOW_RULE_MAX_PRIORITY 0
+
+#define NEXT_ITEM_OF_PATTERN(item, pattern, index)\
+	do {\
+		item = pattern + index;\
+		while (item->type == RTE_FLOW_ITEM_TYPE_VOID) {\
+			index++;\
+			item = pattern + index;\
+		} \
+	} while (0)
+
+#define NEXT_ITEM_OF_ACTION(act, actions, index)\
+	do {\
+		act = actions + index;\
+		while (act->type == RTE_FLOW_ACTION_TYPE_VOID) {\
+			index++;\
+			act = actions + index;\
+		} \
+	} while (0)
+
+/**
+ * Please aware there's an assumption for all the parsers.
+ * rte_flow_item is using big endian, rte_flow_attr and
+ * rte_flow_action are using CPU order.
+ * Because the pattern is used to describe the packets,
+ * normally the packets should use network order.
+ */
+
+/**
+ * Parse the rule to see if it is a n-tuple rule.
+ * And get the n-tuple filter info BTW.
+ * pattern:
+ * The first not void item can be ETH or IPV4.
+ * The second not void item must be IPV4 if the first one is ETH.
+ * The third not void item must be UDP or TCP.
+ * The next not void item must be END.
+ * action:
+ * The first not void action should be QUEUE.
+ * The next not void action should be END.
+ * pattern example:
+ * ITEM		Spec			Mask
+ * ETH		NULL			NULL
+ * IPV4		src_addr 192.168.1.20	0xFFFFFFFF
+ *			dst_addr 192.167.3.50	0xFFFFFFFF
+ *			next_proto_id	17	0xFF
+ * UDP/TCP/	src_port	80	0xFFFF
+ * SCTP		dst_port	80	0xFFFF
+ * END
+ * other members in mask and spec should set to 0x00.
+ * item->last should be NULL.
+ */
+static int
+classify_parse_ntuple_filter(const struct rte_flow_attr *attr,
+			 const struct rte_flow_item pattern[],
+			 const struct rte_flow_action actions[],
+			 struct rte_eth_ntuple_filter *filter,
+			 struct rte_flow_error *error)
+{
+	const struct rte_flow_item *item;
+	const struct rte_flow_action *act;
+	const struct rte_flow_item_ipv4 *ipv4_spec;
+	const struct rte_flow_item_ipv4 *ipv4_mask;
+	const struct rte_flow_item_tcp *tcp_spec;
+	const struct rte_flow_item_tcp *tcp_mask;
+	const struct rte_flow_item_udp *udp_spec;
+	const struct rte_flow_item_udp *udp_mask;
+	const struct rte_flow_item_sctp *sctp_spec;
+	const struct rte_flow_item_sctp *sctp_mask;
+	uint32_t index;
+
+	if (!pattern) {
+		rte_flow_error_set(error,
+			EINVAL, RTE_FLOW_ERROR_TYPE_ITEM_NUM,
+			NULL, "NULL pattern.");
+		return -EINVAL;
+	}
+
+	if (!actions) {
+		rte_flow_error_set(error, EINVAL,
+				   RTE_FLOW_ERROR_TYPE_ACTION_NUM,
+				   NULL, "NULL action.");
+		return -EINVAL;
+	}
+	if (!attr) {
+		rte_flow_error_set(error, EINVAL,
+				   RTE_FLOW_ERROR_TYPE_ATTR,
+				   NULL, "NULL attribute.");
+		return -EINVAL;
+	}
+
+	/* parse pattern */
+	index = 0;
+
+	/* the first not void item can be MAC or IPv4 */
+	NEXT_ITEM_OF_PATTERN(item, pattern, index);
+
+	if (item->type != RTE_FLOW_ITEM_TYPE_ETH &&
+	    item->type != RTE_FLOW_ITEM_TYPE_IPV4) {
+		rte_flow_error_set(error, EINVAL,
+			RTE_FLOW_ERROR_TYPE_ITEM,
+			item, "Not supported by ntuple filter");
+		return -EINVAL;
+	}
+	/* Skip Ethernet */
+	if (item->type == RTE_FLOW_ITEM_TYPE_ETH) {
+		/*Not supported last point for range*/
+		if (item->last) {
+			rte_flow_error_set(error, EINVAL,
+					RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
+					item,
+					"Not supported last point for range");
+			return -EINVAL;
+
+		}
+		/* if the first item is MAC, the content should be NULL */
+		if (item->spec || item->mask) {
+			rte_flow_error_set(error, EINVAL,
+					RTE_FLOW_ERROR_TYPE_ITEM,
+					item,
+					"Not supported by ntuple filter");
+			return -EINVAL;
+		}
+		/* check if the next not void item is IPv4 */
+		index++;
+		NEXT_ITEM_OF_PATTERN(item, pattern, index);
+		if (item->type != RTE_FLOW_ITEM_TYPE_IPV4) {
+			rte_flow_error_set(error, EINVAL,
+					RTE_FLOW_ERROR_TYPE_ITEM,
+					item,
+					"Not supported by ntuple filter");
+			return -EINVAL;
+		}
+	}
+
+	/* get the IPv4 info */
+	if (!item->spec || !item->mask) {
+		rte_flow_error_set(error, EINVAL,
+			RTE_FLOW_ERROR_TYPE_ITEM,
+			item, "Invalid ntuple mask");
+		return -EINVAL;
+	}
+	/*Not supported last point for range*/
+	if (item->last) {
+		rte_flow_error_set(error, EINVAL,
+			RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
+			item, "Not supported last point for range");
+		return -EINVAL;
+
+	}
+
+	ipv4_mask = (const struct rte_flow_item_ipv4 *)item->mask;
+	/**
+	 * Only support src & dst addresses, protocol,
+	 * others should be masked.
+	 */
+	if (ipv4_mask->hdr.version_ihl ||
+		ipv4_mask->hdr.type_of_service ||
+		ipv4_mask->hdr.total_length ||
+		ipv4_mask->hdr.packet_id ||
+		ipv4_mask->hdr.fragment_offset ||
+		ipv4_mask->hdr.time_to_live ||
+		ipv4_mask->hdr.hdr_checksum) {
+		rte_flow_error_set(error,
+			EINVAL, RTE_FLOW_ERROR_TYPE_ITEM,
+			item, "Not supported by ntuple filter");
+		return -EINVAL;
+	}
+
+	filter->dst_ip_mask = ipv4_mask->hdr.dst_addr;
+	filter->src_ip_mask = ipv4_mask->hdr.src_addr;
+	filter->proto_mask  = ipv4_mask->hdr.next_proto_id;
+
+	ipv4_spec = (const struct rte_flow_item_ipv4 *)item->spec;
+	filter->dst_ip = ipv4_spec->hdr.dst_addr;
+	filter->src_ip = ipv4_spec->hdr.src_addr;
+	filter->proto  = ipv4_spec->hdr.next_proto_id;
+
+	/* check if the next not void item is TCP or UDP or SCTP */
+	index++;
+	NEXT_ITEM_OF_PATTERN(item, pattern, index);
+	if (item->type != RTE_FLOW_ITEM_TYPE_TCP &&
+	    item->type != RTE_FLOW_ITEM_TYPE_UDP &&
+	    item->type != RTE_FLOW_ITEM_TYPE_SCTP) {
+		memset(filter, 0, sizeof(struct rte_eth_ntuple_filter));
+		rte_flow_error_set(error, EINVAL,
+			RTE_FLOW_ERROR_TYPE_ITEM,
+			item, "Not supported by ntuple filter");
+		return -EINVAL;
+	}
+
+	/* get the TCP/UDP info */
+	if (!item->spec || !item->mask) {
+		memset(filter, 0, sizeof(struct rte_eth_ntuple_filter));
+		rte_flow_error_set(error, EINVAL,
+			RTE_FLOW_ERROR_TYPE_ITEM,
+			item, "Invalid ntuple mask");
+		return -EINVAL;
+	}
+
+	/*Not supported last point for range*/
+	if (item->last) {
+		memset(filter, 0, sizeof(struct rte_eth_ntuple_filter));
+		rte_flow_error_set(error, EINVAL,
+			RTE_FLOW_ERROR_TYPE_UNSPECIFIED,
+			item, "Not supported last point for range");
+		return -EINVAL;
+
+	}
+
+	if (item->type == RTE_FLOW_ITEM_TYPE_TCP) {
+		tcp_mask = (const struct rte_flow_item_tcp *)item->mask;
+
+		/**
+		 * Only support src & dst ports, tcp flags,
+		 * others should be masked.
+		 */
+		if (tcp_mask->hdr.sent_seq ||
+		    tcp_mask->hdr.recv_ack ||
+		    tcp_mask->hdr.data_off ||
+		    tcp_mask->hdr.rx_win ||
+		    tcp_mask->hdr.cksum ||
+		    tcp_mask->hdr.tcp_urp) {
+			memset(filter, 0,
+				sizeof(struct rte_eth_ntuple_filter));
+			rte_flow_error_set(error, EINVAL,
+				RTE_FLOW_ERROR_TYPE_ITEM,
+				item, "Not supported by ntuple filter");
+			return -EINVAL;
+		}
+
+		filter->dst_port_mask  = tcp_mask->hdr.dst_port;
+		filter->src_port_mask  = tcp_mask->hdr.src_port;
+		if (tcp_mask->hdr.tcp_flags == 0xFF) {
+			filter->flags |= RTE_NTUPLE_FLAGS_TCP_FLAG;
+		} else if (!tcp_mask->hdr.tcp_flags) {
+			filter->flags &= ~RTE_NTUPLE_FLAGS_TCP_FLAG;
+		} else {
+			memset(filter, 0, sizeof(struct rte_eth_ntuple_filter));
+			rte_flow_error_set(error, EINVAL,
+				RTE_FLOW_ERROR_TYPE_ITEM,
+				item, "Not supported by ntuple filter");
+			return -EINVAL;
+		}
+
+		tcp_spec = (const struct rte_flow_item_tcp *)item->spec;
+		filter->dst_port  = tcp_spec->hdr.dst_port;
+		filter->src_port  = tcp_spec->hdr.src_port;
+		filter->tcp_flags = tcp_spec->hdr.tcp_flags;
+	} else if (item->type == RTE_FLOW_ITEM_TYPE_UDP) {
+		udp_mask = (const struct rte_flow_item_udp *)item->mask;
+
+		/**
+		 * Only support src & dst ports,
+		 * others should be masked.
+		 */
+		if (udp_mask->hdr.dgram_len ||
+		    udp_mask->hdr.dgram_cksum) {
+			memset(filter, 0,
+				sizeof(struct rte_eth_ntuple_filter));
+			rte_flow_error_set(error, EINVAL,
+				RTE_FLOW_ERROR_TYPE_ITEM,
+				item, "Not supported by ntuple filter");
+			return -EINVAL;
+		}
+
+		filter->dst_port_mask = udp_mask->hdr.dst_port;
+		filter->src_port_mask = udp_mask->hdr.src_port;
+
+		udp_spec = (const struct rte_flow_item_udp *)item->spec;
+		filter->dst_port = udp_spec->hdr.dst_port;
+		filter->src_port = udp_spec->hdr.src_port;
+	} else {
+		sctp_mask = (const struct rte_flow_item_sctp *)item->mask;
+
+		/**
+		 * Only support src & dst ports,
+		 * others should be masked.
+		 */
+		if (sctp_mask->hdr.tag ||
+		    sctp_mask->hdr.cksum) {
+			memset(filter, 0,
+				sizeof(struct rte_eth_ntuple_filter));
+			rte_flow_error_set(error, EINVAL,
+				RTE_FLOW_ERROR_TYPE_ITEM,
+				item, "Not supported by ntuple filter");
+			return -EINVAL;
+		}
+
+		filter->dst_port_mask = sctp_mask->hdr.dst_port;
+		filter->src_port_mask = sctp_mask->hdr.src_port;
+
+		sctp_spec = (const struct rte_flow_item_sctp *)item->spec;
+		filter->dst_port = sctp_spec->hdr.dst_port;
+		filter->src_port = sctp_spec->hdr.src_port;
+	}
+
+	/* check if the next not void item is END */
+	index++;
+	NEXT_ITEM_OF_PATTERN(item, pattern, index);
+	if (item->type != RTE_FLOW_ITEM_TYPE_END) {
+		memset(filter, 0, sizeof(struct rte_eth_ntuple_filter));
+		rte_flow_error_set(error, EINVAL,
+			RTE_FLOW_ERROR_TYPE_ITEM,
+			item, "Not supported by ntuple filter");
+		return -EINVAL;
+	}
+
+	/* parse action */
+	index = 0;
+
+	/**
+	 * n-tuple only supports count,
+	 * check if the first not void action is COUNT.
+	 */
+	memset(&action, 0, sizeof(action));
+	NEXT_ITEM_OF_ACTION(act, actions, index);
+	if (act->type != RTE_FLOW_ACTION_TYPE_COUNT) {
+		memset(filter, 0, sizeof(struct rte_eth_ntuple_filter));
+		rte_flow_error_set(error, EINVAL,
+			RTE_FLOW_ERROR_TYPE_ACTION,
+			item, "Not supported action.");
+		return -EINVAL;
+	}
+	action.type = RTE_FLOW_ACTION_TYPE_COUNT;
+
+	/* check if the next not void item is END */
+	index++;
+	NEXT_ITEM_OF_ACTION(act, actions, index);
+	if (act->type != RTE_FLOW_ACTION_TYPE_END) {
+		memset(filter, 0, sizeof(struct rte_eth_ntuple_filter));
+		rte_flow_error_set(error, EINVAL,
+			RTE_FLOW_ERROR_TYPE_ACTION,
+			act, "Not supported action.");
+		return -EINVAL;
+	}
+
+	/* parse attr */
+	/* must be input direction */
+	if (!attr->ingress) {
+		memset(filter, 0, sizeof(struct rte_eth_ntuple_filter));
+		rte_flow_error_set(error, EINVAL,
+				   RTE_FLOW_ERROR_TYPE_ATTR_INGRESS,
+				   attr, "Only support ingress.");
+		return -EINVAL;
+	}
+
+	/* not supported */
+	if (attr->egress) {
+		memset(filter, 0, sizeof(struct rte_eth_ntuple_filter));
+		rte_flow_error_set(error, EINVAL,
+				   RTE_FLOW_ERROR_TYPE_ATTR_EGRESS,
+				   attr, "Not support egress.");
+		return -EINVAL;
+	}
+
+	if (attr->priority > 0xFFFF) {
+		memset(filter, 0, sizeof(struct rte_eth_ntuple_filter));
+		rte_flow_error_set(error, EINVAL,
+				   RTE_FLOW_ERROR_TYPE_ATTR_PRIORITY,
+				   attr, "Error priority.");
+		return -EINVAL;
+	}
+	filter->priority = (uint16_t)attr->priority;
+	if (attr->priority >  FLOW_RULE_MIN_PRIORITY)
+		filter->priority = FLOW_RULE_MAX_PRIORITY;
+
+	return 0;
+}
diff --git a/lib/librte_flow_classify/rte_flow_classify_parse.h b/lib/librte_flow_classify/rte_flow_classify_parse.h
new file mode 100644
index 0000000..1d4708a
--- /dev/null
+++ b/lib/librte_flow_classify/rte_flow_classify_parse.h
@@ -0,0 +1,74 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright(c) 2017 Intel Corporation. All rights reserved.
+ *   All rights reserved.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in
+ *       the documentation and/or other materials provided with the
+ *       distribution.
+ *     * Neither the name of Intel Corporation nor the names of its
+ *       contributors may be used to endorse or promote products derived
+ *       from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef _RTE_FLOW_CLASSIFY_PARSE_H_
+#define _RTE_FLOW_CLASSIFY_PARSE_H_
+
+#include <rte_ethdev.h>
+#include <rte_ether.h>
+#include <rte_flow.h>
+#include <stdbool.h>
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+typedef int (*parse_filter_t)(const struct rte_flow_attr *attr,
+			      const struct rte_flow_item pattern[],
+			      const struct rte_flow_action actions[],
+			      struct rte_eth_ntuple_filter *filter,
+			      struct rte_flow_error *error);
+
+/* Skip all VOID items of the pattern */
+void
+classify_pattern_skip_void_item(struct rte_flow_item *items,
+			    const struct rte_flow_item *pattern);
+
+/* Find the first VOID or non-VOID item pointer */
+const struct rte_flow_item *
+classify_find_first_item(const struct rte_flow_item *item, bool is_void);
+
+
+/* Find if there's parse filter function matched */
+parse_filter_t
+classify_find_parse_filter_func(struct rte_flow_item *pattern);
+
+/* get action data */
+struct rte_flow_action *
+classify_get_flow_action(void);
+
+#ifdef __cplusplus
+}
+#endif
+
+#endif /* _RTE_FLOW_CLASSIFY_PARSE_H_ */
diff --git a/lib/librte_flow_classify/rte_flow_classify_version.map b/lib/librte_flow_classify/rte_flow_classify_version.map
new file mode 100644
index 0000000..f7695cb
--- /dev/null
+++ b/lib/librte_flow_classify/rte_flow_classify_version.map
@@ -0,0 +1,12 @@
+EXPERIMENTAL {
+	global:
+
+	rte_flow_classifier_create;
+	rte_flow_classifier_free;
+	rte_flow_classifier_query;
+	rte_flow_classify_table_create;
+	rte_flow_classify_table_entry_add;
+	rte_flow_classify_table_entry_delete;
+
+	local: *;
+};
diff --git a/mk/rte.app.mk b/mk/rte.app.mk
index 8192b98..482656c 100644
--- a/mk/rte.app.mk
+++ b/mk/rte.app.mk
@@ -58,6 +58,7 @@ _LDLIBS-y += -L$(RTE_SDK_BIN)/lib
 #
 # Order is important: from higher level to lower level
 #
+_LDLIBS-$(CONFIG_RTE_LIBRTE_FLOW_CLASSIFY)  += -lrte_flow_classify
 _LDLIBS-$(CONFIG_RTE_LIBRTE_PIPELINE)       += -lrte_pipeline
 _LDLIBS-$(CONFIG_RTE_LIBRTE_TABLE)          += -lrte_table
 _LDLIBS-$(CONFIG_RTE_LIBRTE_PORT)           += -lrte_port
-- 
1.9.1



More information about the dev mailing list