From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mails.dpdk.org (mails.dpdk.org [217.70.189.124]) by inbox.dpdk.org (Postfix) with ESMTP id 4444E43F00; Tue, 30 Apr 2024 18:28:00 +0200 (CEST) Received: from mails.dpdk.org (localhost [127.0.0.1]) by mails.dpdk.org (Postfix) with ESMTP id 5F90C402BF; Tue, 30 Apr 2024 18:27:57 +0200 (CEST) Received: from foss.arm.com (foss.arm.com [217.140.110.172]) by mails.dpdk.org (Postfix) with ESMTP id 0219B400EF for ; Tue, 30 Apr 2024 18:27:54 +0200 (CEST) Received: from usa-sjc-imap-foss1.foss.arm.com (unknown [10.121.207.14]) by usa-sjc-mx-foss1.foss.arm.com (Postfix) with ESMTP id D63AF339; Tue, 30 Apr 2024 09:28:20 -0700 (PDT) Received: from octeon10-1.usa.Arm.com (unknown [10.118.91.161]) by usa-sjc-imap-foss1.foss.arm.com (Postfix) with ESMTPSA id 074E93F73F; Tue, 30 Apr 2024 09:27:53 -0700 (PDT) From: Yoan Picchi To: Thomas Monjalon , Yipeng Wang , Sameh Gobriel , Bruce Richardson , Vladimir Medvedkin Cc: dev@dpdk.org, nd@arm.com, Yoan Picchi , Ruifeng Wang , Nathan Brown Subject: [PATCH v9 1/4] hash: pack the hitmask for hash in bulk lookup Date: Tue, 30 Apr 2024 16:27:40 +0000 Message-Id: <20240430162743.1525484-2-yoan.picchi@arm.com> X-Mailer: git-send-email 2.25.1 In-Reply-To: <20240430162743.1525484-1-yoan.picchi@arm.com> References: <20231020165159.1649282-1-yoan.picchi@arm.com> <20240430162743.1525484-1-yoan.picchi@arm.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Current hitmask includes padding due to Intel's SIMD implementation detail. This patch allows non Intel SIMD implementations to benefit from a dense hitmask. In addition, the new dense hitmask interweave the primary and secondary matches which allow a better cache usage and enable future improvements for the SIMD implementations The default non SIMD path now use this dense mask. Signed-off-by: Yoan Picchi Reviewed-by: Ruifeng Wang Reviewed-by: Nathan Brown --- .mailmap | 2 + lib/hash/arch/arm/compare_signatures.h | 61 +++++++ lib/hash/arch/common/compare_signatures.h | 37 +++++ lib/hash/arch/x86/compare_signatures.h | 53 ++++++ lib/hash/rte_cuckoo_hash.c | 192 ++++++++++++---------- 5 files changed, 254 insertions(+), 91 deletions(-) create mode 100644 lib/hash/arch/arm/compare_signatures.h create mode 100644 lib/hash/arch/common/compare_signatures.h create mode 100644 lib/hash/arch/x86/compare_signatures.h diff --git a/.mailmap b/.mailmap index 66ebc20666..00b50414d3 100644 --- a/.mailmap +++ b/.mailmap @@ -494,6 +494,7 @@ Hari Kumar Vemula Harini Ramakrishnan Hariprasad Govindharajan Harish Patil +Harjot Singh Harman Kalra Harneet Singh Harold Huang @@ -1633,6 +1634,7 @@ Yixue Wang Yi Yang Yi Zhang Yoann Desmouceaux +Yoan Picchi Yogesh Jangra Yogev Chaimovich Yongjie Gu diff --git a/lib/hash/arch/arm/compare_signatures.h b/lib/hash/arch/arm/compare_signatures.h new file mode 100644 index 0000000000..46d15da89f --- /dev/null +++ b/lib/hash/arch/arm/compare_signatures.h @@ -0,0 +1,61 @@ +/* SPDX-License-Identifier: BSD-3-Clause + * Copyright(c) 2010-2016 Intel Corporation + * Copyright(c) 2018-2024 Arm Limited + */ + +/* + * Arm's version uses a densely packed hitmask buffer: + * Every bit is in use. + */ + +#include +#include +#include +#include "rte_cuckoo_hash.h" + +#define DENSE_HASH_BULK_LOOKUP 1 + +static inline void +compare_signatures_dense(uint16_t *hitmask_buffer, + const uint16_t *prim_bucket_sigs, + const uint16_t *sec_bucket_sigs, + uint16_t sig, + enum rte_hash_sig_compare_function sig_cmp_fn) +{ + + static_assert(sizeof(*hitmask_buffer) >= 2 * (RTE_HASH_BUCKET_ENTRIES / 8), + "hitmask_buffer must be wide enough to fit a dense hitmask"); + + /* For match mask every bits indicates the match */ + switch (sig_cmp_fn) { +#if RTE_HASH_BUCKET_ENTRIES <= 8 + case RTE_HASH_COMPARE_NEON: { + uint16x8_t vmat, vsig, x; + int16x8_t shift = {0, 1, 2, 3, 4, 5, 6, 7}; + uint16_t low, high; + + vsig = vld1q_dup_u16((uint16_t const *)&sig); + /* Compare all signatures in the primary bucket */ + vmat = vceqq_u16(vsig, + vld1q_u16((uint16_t const *)prim_bucket_sigs)); + x = vshlq_u16(vandq_u16(vmat, vdupq_n_u16(0x0001)), shift); + low = (uint16_t)(vaddvq_u16(x)); + /* Compare all signatures in the secondary bucket */ + vmat = vceqq_u16(vsig, + vld1q_u16((uint16_t const *)sec_bucket_sigs)); + x = vshlq_u16(vandq_u16(vmat, vdupq_n_u16(0x0001)), shift); + high = (uint16_t)(vaddvq_u16(x)); + *hitmask_buffer = low | high << RTE_HASH_BUCKET_ENTRIES; + + } + break; +#endif + default: + for (unsigned int i = 0; i < RTE_HASH_BUCKET_ENTRIES; i++) { + *hitmask_buffer |= + (sig == prim_bucket_sigs[i]) << i; + *hitmask_buffer |= + ((sig == sec_bucket_sigs[i]) << i) << RTE_HASH_BUCKET_ENTRIES; + } + } +} diff --git a/lib/hash/arch/common/compare_signatures.h b/lib/hash/arch/common/compare_signatures.h new file mode 100644 index 0000000000..f43b367005 --- /dev/null +++ b/lib/hash/arch/common/compare_signatures.h @@ -0,0 +1,37 @@ +/* SPDX-License-Identifier: BSD-3-Clause + * Copyright(c) 2010-2016 Intel Corporation + * Copyright(c) 2018-2024 Arm Limited + */ + +/* + * The generic version could use either a dense or sparsely packed hitmask buffer, + * but the dense one is slightly faster. + */ + +#include +#include +#include +#include "rte_cuckoo_hash.h" + +#define DENSE_HASH_BULK_LOOKUP 1 + +static inline void +compare_signatures_dense(uint16_t *hitmask_buffer, + const uint16_t *prim_bucket_sigs, + const uint16_t *sec_bucket_sigs, + uint16_t sig, + __rte_unused enum rte_hash_sig_compare_function sig_cmp_fn) +{ + + static_assert(sizeof(*hitmask_buffer) >= 2 * (RTE_HASH_BUCKET_ENTRIES / 8), + "hitmask_buffer must be wide enough to fit a dense hitmask"); + + /* For match mask every bits indicates the match */ + for (unsigned int i = 0; i < RTE_HASH_BUCKET_ENTRIES; i++) { + *hitmask_buffer |= + (sig == prim_bucket_sigs[i]) << i; + *hitmask_buffer |= + ((sig == sec_bucket_sigs[i]) << i) << RTE_HASH_BUCKET_ENTRIES; + } + +} diff --git a/lib/hash/arch/x86/compare_signatures.h b/lib/hash/arch/x86/compare_signatures.h new file mode 100644 index 0000000000..db342804e1 --- /dev/null +++ b/lib/hash/arch/x86/compare_signatures.h @@ -0,0 +1,53 @@ +/* SPDX-License-Identifier: BSD-3-Clause + * Copyright(c) 2010-2016 Intel Corporation + * Copyright(c) 2018-2024 Arm Limited + */ + +/* + * x86's version uses a sparsely packed hitmask buffer: + * Every other bit is padding. + */ + +#include +#include +#include +#include "rte_cuckoo_hash.h" + +#define DENSE_HASH_BULK_LOOKUP 0 + +static inline void +compare_signatures_sparse(uint32_t *prim_hash_matches, uint32_t *sec_hash_matches, + const struct rte_hash_bucket *prim_bkt, + const struct rte_hash_bucket *sec_bkt, + uint16_t sig, + enum rte_hash_sig_compare_function sig_cmp_fn) +{ + /* For match mask the first bit of every two bits indicates the match */ + switch (sig_cmp_fn) { +#if defined(__SSE2__) && RTE_HASH_BUCKET_ENTRIES <= 8 + case RTE_HASH_COMPARE_SSE: + /* Compare all signatures in the bucket */ + *prim_hash_matches = _mm_movemask_epi8(_mm_cmpeq_epi16( + _mm_load_si128( + (__m128i const *)prim_bkt->sig_current), + _mm_set1_epi16(sig))); + /* Extract the even-index bits only */ + *prim_hash_matches &= 0x5555; + /* Compare all signatures in the bucket */ + *sec_hash_matches = _mm_movemask_epi8(_mm_cmpeq_epi16( + _mm_load_si128( + (__m128i const *)sec_bkt->sig_current), + _mm_set1_epi16(sig))); + /* Extract the even-index bits only */ + *sec_hash_matches &= 0x5555; + break; +#endif /* defined(__SSE2__) */ + default: + for (unsigned int i = 0; i < RTE_HASH_BUCKET_ENTRIES; i++) { + *prim_hash_matches |= + (sig == prim_bkt->sig_current[i]) << (i << 1); + *sec_hash_matches |= + (sig == sec_bkt->sig_current[i]) << (i << 1); + } + } +} diff --git a/lib/hash/rte_cuckoo_hash.c b/lib/hash/rte_cuckoo_hash.c index 9cf94645f6..0697743cdf 100644 --- a/lib/hash/rte_cuckoo_hash.c +++ b/lib/hash/rte_cuckoo_hash.c @@ -33,6 +33,14 @@ RTE_LOG_REGISTER_DEFAULT(hash_logtype, INFO); #include "rte_cuckoo_hash.h" +#if defined(__ARM_NEON) +#include "arch/arm/compare_signatures.h" +#elif defined(__SSE2__) +#include "arch/x86/compare_signatures.h" +#else +#include "arch/common/compare_signatures.h" +#endif + /* Mask of all flags supported by this version */ #define RTE_HASH_EXTRA_FLAGS_MASK (RTE_HASH_EXTRA_FLAGS_TRANS_MEM_SUPPORT | \ RTE_HASH_EXTRA_FLAGS_MULTI_WRITER_ADD | \ @@ -1857,63 +1865,6 @@ rte_hash_free_key_with_position(const struct rte_hash *h, } -static inline void -compare_signatures(uint32_t *prim_hash_matches, uint32_t *sec_hash_matches, - const struct rte_hash_bucket *prim_bkt, - const struct rte_hash_bucket *sec_bkt, - uint16_t sig, - enum rte_hash_sig_compare_function sig_cmp_fn) -{ - unsigned int i; - - /* For match mask the first bit of every two bits indicates the match */ - switch (sig_cmp_fn) { -#if defined(__SSE2__) - case RTE_HASH_COMPARE_SSE: - /* Compare all signatures in the bucket */ - *prim_hash_matches = _mm_movemask_epi8(_mm_cmpeq_epi16( - _mm_load_si128( - (__m128i const *)prim_bkt->sig_current), - _mm_set1_epi16(sig))); - /* Extract the even-index bits only */ - *prim_hash_matches &= 0x5555; - /* Compare all signatures in the bucket */ - *sec_hash_matches = _mm_movemask_epi8(_mm_cmpeq_epi16( - _mm_load_si128( - (__m128i const *)sec_bkt->sig_current), - _mm_set1_epi16(sig))); - /* Extract the even-index bits only */ - *sec_hash_matches &= 0x5555; - break; -#elif defined(__ARM_NEON) - case RTE_HASH_COMPARE_NEON: { - uint16x8_t vmat, vsig, x; - int16x8_t shift = {-15, -13, -11, -9, -7, -5, -3, -1}; - - vsig = vld1q_dup_u16((uint16_t const *)&sig); - /* Compare all signatures in the primary bucket */ - vmat = vceqq_u16(vsig, - vld1q_u16((uint16_t const *)prim_bkt->sig_current)); - x = vshlq_u16(vandq_u16(vmat, vdupq_n_u16(0x8000)), shift); - *prim_hash_matches = (uint32_t)(vaddvq_u16(x)); - /* Compare all signatures in the secondary bucket */ - vmat = vceqq_u16(vsig, - vld1q_u16((uint16_t const *)sec_bkt->sig_current)); - x = vshlq_u16(vandq_u16(vmat, vdupq_n_u16(0x8000)), shift); - *sec_hash_matches = (uint32_t)(vaddvq_u16(x)); - } - break; -#endif - default: - for (i = 0; i < RTE_HASH_BUCKET_ENTRIES; i++) { - *prim_hash_matches |= - ((sig == prim_bkt->sig_current[i]) << (i << 1)); - *sec_hash_matches |= - ((sig == sec_bkt->sig_current[i]) << (i << 1)); - } - } -} - static inline void __bulk_lookup_l(const struct rte_hash *h, const void **keys, const struct rte_hash_bucket **primary_bkt, @@ -1924,22 +1875,44 @@ __bulk_lookup_l(const struct rte_hash *h, const void **keys, uint64_t hits = 0; int32_t i; int32_t ret; - uint32_t prim_hitmask[RTE_HASH_LOOKUP_BULK_MAX] = {0}; - uint32_t sec_hitmask[RTE_HASH_LOOKUP_BULK_MAX] = {0}; struct rte_hash_bucket *cur_bkt, *next_bkt; +#if DENSE_HASH_BULK_LOOKUP + const int hitmask_padding = 0; + uint16_t hitmask_buffer[RTE_HASH_LOOKUP_BULK_MAX] = {0}; + + static_assert(sizeof(*hitmask_buffer)*8/2 == RTE_HASH_BUCKET_ENTRIES, + "The hitmask must be exactly wide enough to accept the whole hitmask when it is dense"); +#else + const int hitmask_padding = 1; + uint32_t prim_hitmask_buffer[RTE_HASH_LOOKUP_BULK_MAX] = {0}; + uint32_t sec_hitmask_buffer[RTE_HASH_LOOKUP_BULK_MAX] = {0}; +#endif + __hash_rw_reader_lock(h); /* Compare signatures and prefetch key slot of first hit */ for (i = 0; i < num_keys; i++) { - compare_signatures(&prim_hitmask[i], &sec_hitmask[i], +#if DENSE_HASH_BULK_LOOKUP + uint16_t *hitmask = &hitmask_buffer[i]; + compare_signatures_dense(hitmask, + primary_bkt[i]->sig_current, + secondary_bkt[i]->sig_current, + sig[i], h->sig_cmp_fn); + const unsigned int prim_hitmask = *(uint8_t *)(hitmask); + const unsigned int sec_hitmask = *((uint8_t *)(hitmask)+1); +#else + compare_signatures_sparse(&prim_hitmask_buffer[i], &sec_hitmask_buffer[i], primary_bkt[i], secondary_bkt[i], sig[i], h->sig_cmp_fn); + const unsigned int prim_hitmask = prim_hitmask_buffer[i]; + const unsigned int sec_hitmask = sec_hitmask_buffer[i]; +#endif - if (prim_hitmask[i]) { + if (prim_hitmask) { uint32_t first_hit = - rte_ctz32(prim_hitmask[i]) - >> 1; + rte_ctz32(prim_hitmask) + >> hitmask_padding; uint32_t key_idx = primary_bkt[i]->key_idx[first_hit]; const struct rte_hash_key *key_slot = @@ -1950,10 +1923,10 @@ __bulk_lookup_l(const struct rte_hash *h, const void **keys, continue; } - if (sec_hitmask[i]) { + if (sec_hitmask) { uint32_t first_hit = - rte_ctz32(sec_hitmask[i]) - >> 1; + rte_ctz32(sec_hitmask) + >> hitmask_padding; uint32_t key_idx = secondary_bkt[i]->key_idx[first_hit]; const struct rte_hash_key *key_slot = @@ -1967,10 +1940,18 @@ __bulk_lookup_l(const struct rte_hash *h, const void **keys, /* Compare keys, first hits in primary first */ for (i = 0; i < num_keys; i++) { positions[i] = -ENOENT; - while (prim_hitmask[i]) { +#if DENSE_HASH_BULK_LOOKUP + uint16_t *hitmask = &hitmask_buffer[i]; + unsigned int prim_hitmask = *(uint8_t *)(hitmask); + unsigned int sec_hitmask = *((uint8_t *)(hitmask)+1); +#else + unsigned int prim_hitmask = prim_hitmask_buffer[i]; + unsigned int sec_hitmask = sec_hitmask_buffer[i]; +#endif + while (prim_hitmask) { uint32_t hit_index = - rte_ctz32(prim_hitmask[i]) - >> 1; + rte_ctz32(prim_hitmask) + >> hitmask_padding; uint32_t key_idx = primary_bkt[i]->key_idx[hit_index]; const struct rte_hash_key *key_slot = @@ -1992,13 +1973,13 @@ __bulk_lookup_l(const struct rte_hash *h, const void **keys, positions[i] = key_idx - 1; goto next_key; } - prim_hitmask[i] &= ~(3ULL << (hit_index << 1)); + prim_hitmask &= ~(1 << (hit_index << hitmask_padding)); } - while (sec_hitmask[i]) { + while (sec_hitmask) { uint32_t hit_index = - rte_ctz32(sec_hitmask[i]) - >> 1; + rte_ctz32(sec_hitmask) + >> hitmask_padding; uint32_t key_idx = secondary_bkt[i]->key_idx[hit_index]; const struct rte_hash_key *key_slot = @@ -2021,7 +2002,7 @@ __bulk_lookup_l(const struct rte_hash *h, const void **keys, positions[i] = key_idx - 1; goto next_key; } - sec_hitmask[i] &= ~(3ULL << (hit_index << 1)); + sec_hitmask &= ~(1 << (hit_index << hitmask_padding)); } next_key: continue; @@ -2071,11 +2052,20 @@ __bulk_lookup_lf(const struct rte_hash *h, const void **keys, uint64_t hits = 0; int32_t i; int32_t ret; - uint32_t prim_hitmask[RTE_HASH_LOOKUP_BULK_MAX] = {0}; - uint32_t sec_hitmask[RTE_HASH_LOOKUP_BULK_MAX] = {0}; struct rte_hash_bucket *cur_bkt, *next_bkt; uint32_t cnt_b, cnt_a; +#if DENSE_HASH_BULK_LOOKUP + const int hitmask_padding = 0; + uint16_t hitmask_buffer[RTE_HASH_LOOKUP_BULK_MAX] = {0}; + static_assert(sizeof(*hitmask_buffer)*8/2 == RTE_HASH_BUCKET_ENTRIES, + "The hitmask must be exactly wide enough to accept the whole hitmask chen it is dense"); +#else + const int hitmask_padding = 1; + uint32_t prim_hitmask_buffer[RTE_HASH_LOOKUP_BULK_MAX] = {0}; + uint32_t sec_hitmask_buffer[RTE_HASH_LOOKUP_BULK_MAX] = {0}; +#endif + for (i = 0; i < num_keys; i++) positions[i] = -ENOENT; @@ -2089,14 +2079,26 @@ __bulk_lookup_lf(const struct rte_hash *h, const void **keys, /* Compare signatures and prefetch key slot of first hit */ for (i = 0; i < num_keys; i++) { - compare_signatures(&prim_hitmask[i], &sec_hitmask[i], +#if DENSE_HASH_BULK_LOOKUP + uint16_t *hitmask = &hitmask_buffer[i]; + compare_signatures_dense(hitmask, + primary_bkt[i]->sig_current, + secondary_bkt[i]->sig_current, + sig[i], h->sig_cmp_fn); + const unsigned int prim_hitmask = *(uint8_t *)(hitmask); + const unsigned int sec_hitmask = *((uint8_t *)(hitmask)+1); +#else + compare_signatures_sparse(&prim_hitmask_buffer[i], &sec_hitmask_buffer[i], primary_bkt[i], secondary_bkt[i], sig[i], h->sig_cmp_fn); + const unsigned int prim_hitmask = prim_hitmask_buffer[i]; + const unsigned int sec_hitmask = sec_hitmask_buffer[i]; +#endif - if (prim_hitmask[i]) { + if (prim_hitmask) { uint32_t first_hit = - rte_ctz32(prim_hitmask[i]) - >> 1; + rte_ctz32(prim_hitmask) + >> hitmask_padding; uint32_t key_idx = primary_bkt[i]->key_idx[first_hit]; const struct rte_hash_key *key_slot = @@ -2107,10 +2109,10 @@ __bulk_lookup_lf(const struct rte_hash *h, const void **keys, continue; } - if (sec_hitmask[i]) { + if (sec_hitmask) { uint32_t first_hit = - rte_ctz32(sec_hitmask[i]) - >> 1; + rte_ctz32(sec_hitmask) + >> hitmask_padding; uint32_t key_idx = secondary_bkt[i]->key_idx[first_hit]; const struct rte_hash_key *key_slot = @@ -2123,10 +2125,18 @@ __bulk_lookup_lf(const struct rte_hash *h, const void **keys, /* Compare keys, first hits in primary first */ for (i = 0; i < num_keys; i++) { - while (prim_hitmask[i]) { +#if DENSE_HASH_BULK_LOOKUP + uint16_t *hitmask = &hitmask_buffer[i]; + unsigned int prim_hitmask = *(uint8_t *)(hitmask); + unsigned int sec_hitmask = *((uint8_t *)(hitmask)+1); +#else + unsigned int prim_hitmask = prim_hitmask_buffer[i]; + unsigned int sec_hitmask = sec_hitmask_buffer[i]; +#endif + while (prim_hitmask) { uint32_t hit_index = - rte_ctz32(prim_hitmask[i]) - >> 1; + rte_ctz32(prim_hitmask) + >> hitmask_padding; uint32_t key_idx = rte_atomic_load_explicit( &primary_bkt[i]->key_idx[hit_index], @@ -2152,13 +2162,13 @@ __bulk_lookup_lf(const struct rte_hash *h, const void **keys, positions[i] = key_idx - 1; goto next_key; } - prim_hitmask[i] &= ~(3ULL << (hit_index << 1)); + prim_hitmask &= ~(1 << (hit_index << hitmask_padding)); } - while (sec_hitmask[i]) { + while (sec_hitmask) { uint32_t hit_index = - rte_ctz32(sec_hitmask[i]) - >> 1; + rte_ctz32(sec_hitmask) + >> hitmask_padding; uint32_t key_idx = rte_atomic_load_explicit( &secondary_bkt[i]->key_idx[hit_index], @@ -2185,7 +2195,7 @@ __bulk_lookup_lf(const struct rte_hash *h, const void **keys, positions[i] = key_idx - 1; goto next_key; } - sec_hitmask[i] &= ~(3ULL << (hit_index << 1)); + sec_hitmask &= ~(1 << (hit_index << hitmask_padding)); } next_key: continue; -- 2.25.1