From 474b4c1d7cce54d2217d648058ff55d40a92ebd0 Mon Sep 17 00:00:00 2001 From: Ruifeng Wang Date: Mon, 29 Apr 2019 18:02:07 +0800 Subject: [PATCH] hash: simplify signature compare with NEON Replaced multiple neon instructions with single equivalent instruction. This made simpler code and a bit higher performance. Hash bulk lookup had 0.1% ~ 3% performance gain in tests on ARM A72 platforms. Signed-off-by: Ruifeng Wang Reviewed-by: Gavin Hu Acked-by: Yipeng Wang Reviewed-by: Jerin Jacob --- lib/librte_hash/rte_cuckoo_hash.c | 10 +++------- 1 file changed, 3 insertions(+), 7 deletions(-) diff --git a/lib/librte_hash/rte_cuckoo_hash.c b/lib/librte_hash/rte_cuckoo_hash.c index 2dc423fba0..953928f272 100644 --- a/lib/librte_hash/rte_cuckoo_hash.c +++ b/lib/librte_hash/rte_cuckoo_hash.c @@ -1661,7 +1661,6 @@ compare_signatures(uint32_t *prim_hash_matches, uint32_t *sec_hash_matches, #elif defined(RTE_MACHINE_CPUFLAG_NEON) case RTE_HASH_COMPARE_NEON: { uint16x8_t vmat, vsig, x; - uint64x2_t x64; int16x8_t shift = {-15, -13, -11, -9, -7, -5, -3, -1}; vsig = vld1q_dup_u16((uint16_t const *)&sig); @@ -1669,16 +1668,13 @@ compare_signatures(uint32_t *prim_hash_matches, uint32_t *sec_hash_matches, vmat = vceqq_u16(vsig, vld1q_u16((uint16_t const *)prim_bkt->sig_current)); x = vshlq_u16(vandq_u16(vmat, vdupq_n_u16(0x8000)), shift); - x64 = vpaddlq_u32(vpaddlq_u16(x)); - *prim_hash_matches = (uint32_t)(vgetq_lane_u64(x64, 0) + - vgetq_lane_u64(x64, 1)); + *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); - x64 = vpaddlq_u32(vpaddlq_u16(x)); - *sec_hash_matches = (uint32_t)(vgetq_lane_u64(x64, 0) + - vgetq_lane_u64(x64, 1)); } + *sec_hash_matches = (uint32_t)(vaddvq_u16(x)); + } break; #endif default: -- 2.20.1