efd: support lookup using NEON intrinsics

* Added file lib/librte_efd/rte_efd_arm64.h to hold arm64
  specific definitions
* Verified the changes with efd_autotest unit test case

Signed-off-by: Ashwin Sekhar T K <ashwin.sekhar@caviumnetworks.com>

MAINTAINERS

@@ -147,6 +147,7 @@ F: lib/librte_eal/common/include/arch/arm/*_64.h
 F: lib/librte_acl/acl_run_neon.*
 F: lib/librte_lpm/rte_lpm_neon.h
 F: lib/librte_hash/rte*_arm64.h
+F: lib/librte_efd/rte*_arm64.h
 F: drivers/net/ixgbe/ixgbe_rxtx_vec_neon.c
 F: drivers/net/i40e/i40e_rxtx_vec_neon.c
 F: drivers/net/virtio/virtio_rxtx_simple_neon.c

lib/librte_efd/rte_efd.c

@@ -53,6 +53,8 @@
 #include "rte_efd.h"
 #if defined(RTE_ARCH_X86)
 #include "rte_efd_x86.h"
+#elif defined(RTE_ARCH_ARM64)
+#include "rte_efd_arm64.h"
 #endif
 
 #define EFD_KEY(key_idx, table) (table->keys + ((key_idx) * table->key_len))
@@ -103,6 +105,7 @@ allocated memory
 enum efd_lookup_internal_function {
 	EFD_LOOKUP_SCALAR = 0,
 	EFD_LOOKUP_AVX2,
+	EFD_LOOKUP_NEON,
 	EFD_LOOKUP_NUM
 };
 
@@ -673,6 +676,16 @@ rte_efd_create(const char *name, uint32_t max_num_rules, uint32_t key_len,
 	if (RTE_EFD_VALUE_NUM_BITS > 3 && rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2))
 		table->lookup_fn = EFD_LOOKUP_AVX2;
 	else
+#endif
+#if defined(RTE_ARCH_ARM64)
+	/*
+	 * For less than or equal to 16 bits, scalar function performs better
+	 * than vectorised version
+	 */
+	if (RTE_EFD_VALUE_NUM_BITS > 16 &&
+	    rte_cpu_get_flag_enabled(RTE_CPUFLAG_NEON))
+		table->lookup_fn = EFD_LOOKUP_NEON;
+	else
 #endif
 		table->lookup_fn = EFD_LOOKUP_SCALAR;
 
@@ -1271,6 +1284,15 @@ efd_lookup_internal(const struct efd_online_group_entry * const group,
 					group->lookup_table,
 					hash_val_a,
 					hash_val_b);
+		break;
+#endif
+#if defined(RTE_ARCH_ARM64)
+	case EFD_LOOKUP_NEON:
+		return efd_lookup_internal_neon(group->hash_idx,
+					group->lookup_table,
+					hash_val_a,
+					hash_val_b);
+		break;
 #endif
 	case EFD_LOOKUP_SCALAR:
 	/* Fall-through */

lib/librte_efd/rte_efd_arm64.h

@@ -0,0 +1,76 @@
+/*
+ *   BSD LICENSE
+ *
+ *   Copyright (C) Cavium networks Ltd. 2017.
+ *
+ *   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 Cavium networks 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.
+ */
+
+/*
+ * rte_efd_arm64.h
+ * This file holds all arm64 specific EFD functions
+ */
+
+#ifndef __RTE_EFD_ARM64_H__
+#define __RTE_EFD_ARM64_H__
+
+#include <rte_vect.h>
+
+static inline efd_value_t
+efd_lookup_internal_neon(const efd_hashfunc_t *group_hash_idx,
+		const efd_lookuptbl_t *group_lookup_table,
+		const uint32_t hash_val_a, const uint32_t hash_val_b)
+{
+	efd_value_t value = 0;
+	uint32_t i = 0;
+	uint32x4_t vhash_val_a = vmovq_n_u32(hash_val_a);
+	uint32x4_t vhash_val_b = vmovq_n_u32(hash_val_b);
+	int32x4_t vshift = {0, 1, 2, 3};
+	uint32x4_t vmask = vdupq_n_u32(0x1);
+	int32x4_t vincr = vdupq_n_s32(4);
+
+	for (; i < RTE_EFD_VALUE_NUM_BITS; i += 4) {
+		uint32x4_t vhash_idx = vshll_n_u16(
+			vld1_u16((uint16_t const *)&group_hash_idx[i]), 0);
+		uint32x4_t vlookup_table = vshll_n_u16(
+			vld1_u16((uint16_t const *)&group_lookup_table[i]), 0);
+		uint32x4_t vhash = vaddq_u32(vhash_val_a,
+					vmulq_u32(vhash_idx, vhash_val_b));
+		int32x4_t vbucket_idx = vnegq_s32(vreinterpretq_s32_u32(
+				vshrq_n_u32(vhash, EFD_LOOKUPTBL_SHIFT)));
+		uint32x4_t vresult = vshlq_u32(vlookup_table, vbucket_idx);
+
+		vresult = vandq_u32(vresult, vmask);
+		vresult = vshlq_u32(vresult, vshift);
+		value |= vaddvq_u32(vresult);
+		vshift = vaddq_s32(vshift, vincr);
+	}
+
+	return value;
+}
+
+#endif /* __RTE_EFD_ARM64_H__ */