From 744b13e302559fc6da30749df266810882b5df99 Mon Sep 17 00:00:00 2001
From: mwish <maplewish117@gmail.com>
Date: Wed, 28 Aug 2024 11:44:32 +0800
Subject: [PATCH] Using intrinsics to optimize counting HyperLogLog trailing
 bits (#846)

Godbolt link: https://godbolt.org/z/3YPvxsr5s

__builtin_ctz would generate shorter code than hand-written loop.

---------

Signed-off-by: mwish <maplewish117@gmail.com>
Signed-off-by: Binbin <binloveplay1314@qq.com>
Signed-off-by: Madelyn Olson <madelyneolson@gmail.com>
Co-authored-by: Binbin <binloveplay1314@qq.com>
Co-authored-by: Madelyn Olson <madelyneolson@gmail.com>
---
 src/hyperloglog.c | 19 ++++++-------------
 src/intrinsics.h  | 21 +++++++++++++++++++++
 2 files changed, 27 insertions(+), 13 deletions(-)
 create mode 100644 src/intrinsics.h

diff --git a/src/hyperloglog.c b/src/hyperloglog.c
index 9769533d5..79d81ac8f 100644
--- a/src/hyperloglog.c
+++ b/src/hyperloglog.c
@@ -30,6 +30,7 @@
  */
 
 #include "server.h"
+#include "intrinsics.h"
 
 #include <stdint.h>
 #include <math.h>
@@ -455,7 +456,7 @@ uint64_t MurmurHash64A(const void *key, int len, unsigned int seed) {
  * of the pattern 000..1 of the element hash. As a side effect 'regp' is
  * set to the register index this element hashes to. */
 int hllPatLen(unsigned char *ele, size_t elesize, long *regp) {
-    uint64_t hash, bit, index;
+    uint64_t hash, index;
     int count;
 
     /* Count the number of zeroes starting from bit HLL_REGISTERS
@@ -465,21 +466,13 @@ int hllPatLen(unsigned char *ele, size_t elesize, long *regp) {
      * Note that the final "1" ending the sequence of zeroes must be
      * included in the count, so if we find "001" the count is 3, and
      * the smallest count possible is no zeroes at all, just a 1 bit
-     * at the first position, that is a count of 1.
-     *
-     * This may sound like inefficient, but actually in the average case
-     * there are high probabilities to find a 1 after a few iterations. */
+     * at the first position, that is a count of 1. */
     hash = MurmurHash64A(ele, elesize, 0xadc83b19ULL);
     index = hash & HLL_P_MASK;      /* Register index. */
     hash >>= HLL_P;                 /* Remove bits used to address the register. */
-    hash |= ((uint64_t)1 << HLL_Q); /* Make sure the loop terminates
-                                       and count will be <= Q+1. */
-    bit = 1;
-    count = 1; /* Initialized to 1 since we count the "00000...1" pattern. */
-    while ((hash & bit) == 0) {
-        count++;
-        bit <<= 1;
-    }
+    hash |= ((uint64_t)1 << HLL_Q); /* Make sure count will be <= Q+1. */
+    count = 1;                      /* Initialized to 1 since we count the "00000...1" pattern. */
+    count += builtin_ctzll(hash);
     *regp = (int)index;
     return count;
 }
diff --git a/src/intrinsics.h b/src/intrinsics.h
new file mode 100644
index 000000000..3be419915
--- /dev/null
+++ b/src/intrinsics.h
@@ -0,0 +1,21 @@
+#ifndef __INTRINSICS_H
+#define __INTRINSICS_H
+
+#include <stdint.h>
+
+/* Count the number of trailing zero bits in a 64-bit integer. */
+static inline int32_t builtin_ctzll(uint64_t value) {
+    if (value == 0) return 64;
+#if defined(__clang__) || defined(__GNUC__)
+    return __builtin_ctzll(value);
+#else
+    int bitpos = 0;
+    while (value & 1 == 0) {
+        value >>= 1;
+        ++bitpos;
+    }
+    return bitpos;
+#endif
+}
+
+#endif /* __INTRINSICS_H */