summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorRuifeng Wang <ruifeng.wang@arm.com>2019-04-29 18:02:07 +0800
committerThomas Monjalon <thomas@monjalon.net>2019-06-05 19:49:28 +0200
commit474b4c1d7cce54d2217d648058ff55d40a92ebd0 (patch)
treeafb48cdc114f4c6ca130a11194ac706dcca0f298
parent4c773788e05a3d5e2a55e134f015bea09350807b (diff)
downloaddpdk-next-eventdev-474b4c1d7cce54d2217d648058ff55d40a92ebd0.zip
dpdk-next-eventdev-474b4c1d7cce54d2217d648058ff55d40a92ebd0.tar.gz
dpdk-next-eventdev-474b4c1d7cce54d2217d648058ff55d40a92ebd0.tar.xz
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 <ruifeng.wang@arm.com> Reviewed-by: Gavin Hu <gavin.hu@arm.com> Acked-by: Yipeng Wang <yipeng1.wang@intel.com> Reviewed-by: Jerin Jacob <jerinj@marvell.com>
-rw-r--r--lib/librte_hash/rte_cuckoo_hash.c10
1 files changed, 3 insertions, 7 deletions
diff --git a/lib/librte_hash/rte_cuckoo_hash.c b/lib/librte_hash/rte_cuckoo_hash.c
index 2dc423f..953928f 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: