From efae4bfd1357d0f81d3a04fee7bb1914a0a2401e Mon Sep 17 00:00:00 2001
From: Daniel Lemire <daniel@lemire.me>
Date: Tue, 21 Nov 2023 10:42:02 -0500
Subject: [PATCH 1/4] src: implement FastByteLengthUtf8 with
 simdutf::utf8_length_from_latin1

---
 src/node_buffer.cc | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/src/node_buffer.cc b/src/node_buffer.cc
index ff041274f90d24..ab7647bc87b197 100644
--- a/src/node_buffer.cc
+++ b/src/node_buffer.cc
@@ -743,14 +743,7 @@ void SlowByteLengthUtf8(const FunctionCallbackInfo<Value>& args) {
 
 uint32_t FastByteLengthUtf8(Local<Value> receiver,
                             const v8::FastOneByteString& source) {
-  uint32_t result = 0;
-  uint32_t length = source.length;
-  const uint8_t* data = reinterpret_cast<const uint8_t*>(source.data);
-  for (uint32_t i = 0; i < length; ++i) {
-    result += (data[i] >> 7);
-  }
-  result += length;
-  return result;
+  return simdutf::utf8_length_from_latin1(source.data, source.length);
 }
 
 static v8::CFunction fast_byte_length_utf8(

From 1eb654e274ce48c2ada2cba6f93e07b554222b1f Mon Sep 17 00:00:00 2001
From: Daniel Lemire <dlemire@lemire.me>
Date: Sun, 26 Nov 2023 17:10:50 -0500
Subject: [PATCH 2/4] guard simdutf usage and adds latin1 string to bench

---
 benchmark/buffers/buffer-bytelength-string.js |  3 ++-
 src/node_buffer.cc                            | 13 ++++++++++++-
 2 files changed, 14 insertions(+), 2 deletions(-)

diff --git a/benchmark/buffers/buffer-bytelength-string.js b/benchmark/buffers/buffer-bytelength-string.js
index fc0c005e7f9e6a..9f5c3f744c7222 100644
--- a/benchmark/buffers/buffer-bytelength-string.js
+++ b/benchmark/buffers/buffer-bytelength-string.js
@@ -2,7 +2,7 @@
 const common = require('../common');
 
 const bench = common.createBenchmark(main, {
-  type: ['one_byte', 'two_bytes', 'three_bytes', 'four_bytes'],
+  type: ['one_byte', 'two_bytes', 'three_bytes', 'four_bytes', 'latin1'],
   encoding: ['utf8', 'base64'],
   repeat: [1, 2, 16, 256], // x16
   n: [4e6],
@@ -14,6 +14,7 @@ const chars = {
   two_bytes: 'ΰαβγδεζηθικλμνξο',
   three_bytes: '挰挱挲挳挴挵挶挷挸挹挺挻挼挽挾挿',
   four_bytes: '𠜎𠜱𠝹𠱓𠱸𠲖𠳏𠳕𠴕𠵼𠵿𠸎𠸏𠹷𠺝𠺢',
+  latin1: 'Un homme sage est supérieur à toutes les insultes qui peuvent lui être adressées, et la meilleure réponse est la patience et la modération.'
 };
 
 function getInput(type, repeat, encoding) {
diff --git a/src/node_buffer.cc b/src/node_buffer.cc
index ab7647bc87b197..02470d07bcb27e 100644
--- a/src/node_buffer.cc
+++ b/src/node_buffer.cc
@@ -743,7 +743,18 @@ void SlowByteLengthUtf8(const FunctionCallbackInfo<Value>& args) {
 
 uint32_t FastByteLengthUtf8(Local<Value> receiver,
                             const v8::FastOneByteString& source) {
-  return simdutf::utf8_length_from_latin1(source.data, source.length);
+  // For short inputs, the function call overhead to simdutf is maybe
+  // not worth it, reserve simdutf for long strings.
+  if(source.length > 128) {
+    return simdutf::utf8_length_from_latin1(source.data, source.length);
+  }
+  uint32_t length = source.length;
+  uint32_t result = length;
+  const uint8_t* data = reinterpret_cast<const uint8_t*>(source.data);
+  for (uint32_t i = 0; i < length; ++i) {
+    result += (data[i] >> 7);
+  }
+  return result;
 }
 
 static v8::CFunction fast_byte_length_utf8(

From 074dfbbe5e8ed53aae275d65e59ec80197b2280f Mon Sep 17 00:00:00 2001
From: Daniel Lemire <dlemire@lemire.me>
Date: Mon, 27 Nov 2023 12:10:20 -0500
Subject: [PATCH 3/4] lint

---
 src/node_buffer.cc | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/node_buffer.cc b/src/node_buffer.cc
index 02470d07bcb27e..300060f9d24290 100644
--- a/src/node_buffer.cc
+++ b/src/node_buffer.cc
@@ -745,7 +745,7 @@ uint32_t FastByteLengthUtf8(Local<Value> receiver,
                             const v8::FastOneByteString& source) {
   // For short inputs, the function call overhead to simdutf is maybe
   // not worth it, reserve simdutf for long strings.
-  if(source.length > 128) {
+  if (source.length > 128) {
     return simdutf::utf8_length_from_latin1(source.data, source.length);
   }
   uint32_t length = source.length;

From eb021106099319f2b538aebdb8996a51a3a8ceff Mon Sep 17 00:00:00 2001
From: Daniel Lemire <dlemire@lemire.me>
Date: Mon, 4 Dec 2023 18:13:53 -0500
Subject: [PATCH 4/4] fix: lint

---
 benchmark/buffers/buffer-bytelength-string.js | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/benchmark/buffers/buffer-bytelength-string.js b/benchmark/buffers/buffer-bytelength-string.js
index 9f5c3f744c7222..143da0215a613b 100644
--- a/benchmark/buffers/buffer-bytelength-string.js
+++ b/benchmark/buffers/buffer-bytelength-string.js
@@ -2,7 +2,8 @@
 const common = require('../common');
 
 const bench = common.createBenchmark(main, {
-  type: ['one_byte', 'two_bytes', 'three_bytes', 'four_bytes', 'latin1'],
+  type: ['one_byte', 'two_bytes', 'three_bytes',
+         'four_bytes', 'latin1'],
   encoding: ['utf8', 'base64'],
   repeat: [1, 2, 16, 256], // x16
   n: [4e6],
@@ -14,7 +15,8 @@ const chars = {
   two_bytes: 'ΰαβγδεζηθικλμνξο',
   three_bytes: '挰挱挲挳挴挵挶挷挸挹挺挻挼挽挾挿',
   four_bytes: '𠜎𠜱𠝹𠱓𠱸𠲖𠳏𠳕𠴕𠵼𠵿𠸎𠸏𠹷𠺝𠺢',
-  latin1: 'Un homme sage est supérieur à toutes les insultes qui peuvent lui être adressées, et la meilleure réponse est la patience et la modération.'
+  latin1: 'Un homme sage est supérieur à toutes ' +
+    'les insultes qui peuvent lui être adressées, et la meilleure réponse est la patience et la modération.',
 };
 
 function getInput(type, repeat, encoding) {