DPDK patches and discussions
 help / color / mirror / Atom feed
* [dpdk-dev] [PATCH v1] hash: simplify signature compare neon process
@ 2019-04-29 10:02 Ruifeng Wang
  2019-04-29 10:02 ` Ruifeng Wang
                   ` (2 more replies)
  0 siblings, 3 replies; 7+ messages in thread
From: Ruifeng Wang @ 2019-04-29 10:02 UTC (permalink / raw)
  To: yipeng1.wang, sameh.gobriel, bruce.richardson, pablo.de.lara.guarch
  Cc: dev, jerinj, Honnappa.Nagarahalli, nd, Ruifeng Wang

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>
---
 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 261267b..f17819e 100644
--- a/lib/librte_hash/rte_cuckoo_hash.c
+++ b/lib/librte_hash/rte_cuckoo_hash.c
@@ -1656,7 +1656,6 @@ 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);
@@ -1664,16 +1663,13 @@ 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:
-- 
2.7.4

^ permalink raw reply	[flat|nested] 7+ messages in thread
* Re: [dpdk-dev] [PATCH v1] hash: simplify signature compare neon process
@ 2019-06-04 16:58 Jerin Jacob Kollanukkaran
  0 siblings, 0 replies; 7+ messages in thread
From: Jerin Jacob Kollanukkaran @ 2019-06-04 16:58 UTC (permalink / raw)
  To: Ruifeng Wang, yipeng1.wang, sameh.gobriel, bruce.richardson,
	pablo.de.lara.guarch
  Cc: dev, Honnappa.Nagarahalli, nd

> -----Original Message-----
> From: Ruifeng Wang <ruifeng.wang@arm.com>
> Sent: Monday, April 29, 2019 3:32 PM
> To: yipeng1.wang@intel.com; sameh.gobriel@intel.com;
> bruce.richardson@intel.com; pablo.de.lara.guarch@intel.com
> Cc: dev@dpdk.org; Jerin Jacob Kollanukkaran <jerinj@marvell.com>;
> Honnappa.Nagarahalli@arm.com; nd@arm.com; Ruifeng Wang
> <ruifeng.wang@arm.com>
> Subject: [EXT] [PATCH v1] hash: simplify signature compare neon process
> 
> 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>

Reviewed-by: Jerin Jacob <jerinj@marvell.com>

^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2019-06-05 17:25 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-04-29 10:02 [dpdk-dev] [PATCH v1] hash: simplify signature compare neon process Ruifeng Wang
2019-04-29 10:02 ` Ruifeng Wang
2019-05-03 20:40 ` Thomas Monjalon
2019-05-03 20:40   ` Thomas Monjalon
2019-05-22  0:28 ` Wang, Yipeng1
2019-06-05 17:25   ` Thomas Monjalon
2019-06-04 16:58 Jerin Jacob Kollanukkaran

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).