From eb4038fa83009451d2ad9a1eda4cd6bcdf613958 Mon Sep 17 00:00:00 2001
From: Aliaksandr Kalenik <kalenik.aliaksandr@gmail.com>
Date: Wed, 22 Apr 2026 15:02:57 +0200
Subject: [PATCH] AK: Fix Utf16View::operator<=> code-unit ordering on
 little-endian

The !has_ascii_storage() && !other.has_ascii_storage() branch did a
byte-wise __builtin_memcmp over a char16_t array, which on little-endian
does not give code-unit order: the low byte is compared first, so
0xD83D (bytes [0x3D, 0xD8]) spuriously compared less than 0x2764
(bytes [0x64, 0x27]) even though the code unit 0xD83D is greater.

No in-tree caller currently uses operator<=> for Utf16View ordering,
so this bug is dormant; the follow-up LibJS change exposes it.

Replace the memcmp branch with a per-code-unit loop, which the compiler
can auto-vectorize and which mirrors what is_code_unit_less_than already
does.
---
 AK/Utf16View.h             |  2 --
 Tests/AK/TestUtf16View.cpp | 10 ++++++++++
 2 files changed, 10 insertions(+), 2 deletions(-)

diff --git a/AK/Utf16View.h b/AK/Utf16View.h
index fadc5bd504..3660174a19 100644
--- a/AK/Utf16View.h
+++ b/AK/Utf16View.h
@@ -261,8 +261,6 @@ public:
 
         if (has_ascii_storage() && other.has_ascii_storage()) {
             result = __builtin_memcmp(m_string.ascii, other.m_string.ascii, length);
-        } else if (!has_ascii_storage() && !other.has_ascii_storage()) {
-            result = __builtin_memcmp(m_string.utf16, other.m_string.utf16, length * sizeof(char16_t));
         } else {
             for (size_t i = 0; i < length; ++i) {
                 auto this_code_unit = code_unit_at(i);
diff --git a/Tests/AK/TestUtf16View.cpp b/Tests/AK/TestUtf16View.cpp
index 1bcf606453..87759127bb 100644
--- a/Tests/AK/TestUtf16View.cpp
+++ b/Tests/AK/TestUtf16View.cpp
@@ -482,6 +482,16 @@ TEST_CASE(comparison)
     EXPECT(u"😂"sv > u"😀"sv);
     EXPECT(!(u"😂"sv <= u"😀"sv));
     EXPECT(u"😂"sv >= u"😀"sv);
+
+    EXPECT(u"ÿ"sv < u"Ā"sv);
+    EXPECT(!(u"ÿ"sv > u"Ā"sv));
+    EXPECT(u"Ā"sv > u"ÿ"sv);
+    EXPECT(!(u"Ā"sv < u"ÿ"sv));
+
+    EXPECT(u"❤"sv < u"😀"sv);
+    EXPECT(!(u"❤"sv > u"😀"sv));
+    EXPECT(u"😀"sv > u"❤"sv);
+    EXPECT(!(u"😀"sv < u"❤"sv));
 }
 
 TEST_CASE(equals_ignoring_case)