From 5f973a39d67a744d514ee80e05a1c7f40bc0ebc6 Mon Sep 17 00:00:00 2001
From: Nathan Bossart <nathandbossart@gmail.com>
Date: Thu, 25 Aug 2022 22:18:30 -0700
Subject: [PATCH v3 1/2] abstract architecture-specific implementation details
 from pg_lfind32()

---
 src/include/port/pg_lfind.h | 55 ++++++++++++++++++----------------
 src/include/port/simd.h     | 60 +++++++++++++++++++++++++++++++++++++
 2 files changed, 90 insertions(+), 25 deletions(-)

diff --git a/src/include/port/pg_lfind.h b/src/include/port/pg_lfind.h
index a4e13dffec..7a851ea42c 100644
--- a/src/include/port/pg_lfind.h
+++ b/src/include/port/pg_lfind.h
@@ -91,16 +91,19 @@ pg_lfind32(uint32 key, uint32 *base, uint32 nelem)
 {
 	uint32		i = 0;
 
-#ifdef USE_SSE2
+#ifndef USE_NO_SIMD
 
 	/*
-	 * A 16-byte register only has four 4-byte lanes. For better
-	 * instruction-level parallelism, each loop iteration operates on a block
-	 * of four registers. Testing has showed this is ~40% faster than using a
-	 * block of two registers.
+	 * For better instruction-level parallelism, each loop iteration operates
+	 * on a block of four registers.  Testing for SSE2 has showed this is ~40%
+	 * faster than using a block of two registers.
 	 */
-	const		__m128i keys = _mm_set1_epi32(key); /* load 4 copies of key */
-	uint32		iterations = nelem & ~0xF;	/* round down to multiple of 16 */
+	const Vector32 keys = vector32_broadcast(key);	/* load copies of key */
+	uint32		nelem_per_vector = sizeof(Vector32) / sizeof(uint32);
+	uint32		nelem_per_iteration = 4 * nelem_per_vector;
+
+	/* round down to multiple of elements per iteration */
+	uint32		tail_idx = nelem & ~(nelem_per_iteration - 1);
 
 #if defined(USE_ASSERT_CHECKING)
 	bool		assert_result = false;
@@ -116,31 +119,33 @@ pg_lfind32(uint32 key, uint32 *base, uint32 nelem)
 	}
 #endif
 
-	for (i = 0; i < iterations; i += 16)
+	for (i = 0; i < tail_idx; i += nelem_per_iteration)
 	{
-		/* load the next block into 4 registers holding 4 values each */
-		const		__m128i vals1 = _mm_loadu_si128((__m128i *) & base[i]);
-		const		__m128i vals2 = _mm_loadu_si128((__m128i *) & base[i + 4]);
-		const		__m128i vals3 = _mm_loadu_si128((__m128i *) & base[i + 8]);
-		const		__m128i vals4 = _mm_loadu_si128((__m128i *) & base[i + 12]);
+		Vector32	vals1, vals2, vals3, vals4,
+					result1, result2, result3, result4,
+					tmp1, tmp2, result;
+
+		/* load the next block into 4 registers */
+		vector32_load(&vals1, &base[i]);
+		vector32_load(&vals2, &base[i + nelem_per_vector]);
+		vector32_load(&vals3, &base[i + nelem_per_vector * 2]);
+		vector32_load(&vals4, &base[i + nelem_per_vector * 3]);
 
 		/* compare each value to the key */
-		const		__m128i result1 = _mm_cmpeq_epi32(keys, vals1);
-		const		__m128i result2 = _mm_cmpeq_epi32(keys, vals2);
-		const		__m128i result3 = _mm_cmpeq_epi32(keys, vals3);
-		const		__m128i result4 = _mm_cmpeq_epi32(keys, vals4);
+		result1 = vector32_eq(keys, vals1);
+		result2 = vector32_eq(keys, vals2);
+		result3 = vector32_eq(keys, vals3);
+		result4 = vector32_eq(keys, vals4);
 
 		/* combine the results into a single variable */
-		const		__m128i tmp1 = _mm_or_si128(result1, result2);
-		const		__m128i tmp2 = _mm_or_si128(result3, result4);
-		const		__m128i result = _mm_or_si128(tmp1, tmp2);
+		tmp1 = vector32_or(result1, result2);
+		tmp2 = vector32_or(result3, result4);
+		result = vector32_or(tmp1, tmp2);
 
 		/* see if there was a match */
-		if (_mm_movemask_epi8(result) != 0)
+		if (vector32_any_lane_set(result))
 		{
-#if defined(USE_ASSERT_CHECKING)
 			Assert(assert_result == true);
-#endif
 			return true;
 		}
 	}
@@ -151,14 +156,14 @@ pg_lfind32(uint32 key, uint32 *base, uint32 nelem)
 	{
 		if (key == base[i])
 		{
-#if defined(USE_SSE2) && defined(USE_ASSERT_CHECKING)
+#ifndef USE_NO_SIMD
 			Assert(assert_result == true);
 #endif
 			return true;
 		}
 	}
 
-#if defined(USE_SSE2) && defined(USE_ASSERT_CHECKING)
+#ifndef USE_NO_SIMD
 	Assert(assert_result == false);
 #endif
 	return false;
diff --git a/src/include/port/simd.h b/src/include/port/simd.h
index 8f85153110..bd4f1a3f39 100644
--- a/src/include/port/simd.h
+++ b/src/include/port/simd.h
@@ -32,6 +32,7 @@
 #include <emmintrin.h>
 #define USE_SSE2
 typedef __m128i Vector8;
+typedef __m128i Vector32;
 
 #else
 /*
@@ -40,18 +41,24 @@ typedef __m128i Vector8;
  */
 #define USE_NO_SIMD
 typedef uint64 Vector8;
+typedef uint64 Vector32;
 #endif
 
 
 static inline void vector8_load(Vector8 *v, const uint8 *s);
+static inline void vector32_load(Vector32 *v, const uint32 *s);
 static inline Vector8 vector8_broadcast(const uint8 c);
+static inline Vector32 vector32_broadcast(const uint32 c);
 static inline bool vector8_has_zero(const Vector8 v);
 static inline bool vector8_has(const Vector8 v, const uint8 c);
 static inline bool vector8_has_le(const Vector8 v, const uint8 c);
 static inline bool vector8_is_highbit_set(const Vector8 v);
 static inline Vector8 vector8_or(const Vector8 v1, const Vector8 v2);
+static inline Vector32 vector32_or(const Vector32 v1, const Vector32 v2);
 #ifndef USE_NO_SIMD
 static inline Vector8 vector8_eq(const Vector8 v1, const Vector8 v2);
+static inline Vector32 vector32_eq(const Vector32 v1, const Vector32 v2);
+static inline bool vector32_any_lane_set(const Vector32 v);
 #endif
 
 /*
@@ -68,6 +75,16 @@ vector8_load(Vector8 *v, const uint8 *s)
 #endif
 }
 
+static inline void
+vector32_load(Vector32 *v, const uint32 *s)
+{
+#ifdef USE_SSE2
+	*v = _mm_loadu_si128((const __m128i *) s);
+#else
+	memcpy(v, s, sizeof(Vector32));
+#endif
+}
+
 
 /*
  * Functions for creating a vector with all elements set to the same value.
@@ -83,6 +100,16 @@ vector8_broadcast(const uint8 c)
 #endif
 }
 
+static inline Vector32
+vector32_broadcast(const uint32 c)
+{
+#ifdef USE_SSE2
+	return _mm_set1_epi32(c);
+#else
+	return ~UINT64CONST(0) / 0xFFFFFFFF * c;
+#endif
+}
+
 
 /*
  * Functions for comparing vector elements to a given value.
@@ -196,6 +223,21 @@ vector8_is_highbit_set(const Vector8 v)
 #endif
 }
 
+/*
+ * vector32_any_lane_set() assumes that each lane in the given vector is set to
+ * either all ones or all zeroes.  If this is not true, its behavior is
+ * undefined.
+ */
+#ifndef USE_NO_SIMD
+static inline bool
+vector32_any_lane_set(const Vector32 v)
+{
+#ifdef USE_SSE2
+	return _mm_movemask_epi8(v) != 0;
+#endif
+}
+#endif
+
 /* comparisons between vectors */
 
 #ifndef USE_NO_SIMD
@@ -206,6 +248,14 @@ vector8_eq(const Vector8 v1, const Vector8 v2)
 	return _mm_cmpeq_epi8(v1, v2);
 #endif
 }
+
+static inline Vector32
+vector32_eq(const Vector32 v1, const Vector32 v2)
+{
+#ifdef USE_SSE2
+	return _mm_cmpeq_epi32(v1, v2);
+#endif
+}
 #endif
 
 /* bitwise operations */
@@ -220,4 +270,14 @@ vector8_or(const Vector8 v1, const Vector8 v2)
 #endif
 }
 
+static inline Vector32
+vector32_or(const Vector32 v1, const Vector32 v2)
+{
+#ifdef USE_SSE2
+	return _mm_or_si128(v1, v2);
+#else
+	return v1 | v2;
+#endif
+}
+
 #endif							/* SIMD_H */
-- 
2.25.1

