From 206479b2b5fc2641a619eb0d05c1185d869ef844 Mon Sep 17 00:00:00 2001
From: Andreas Kling <andreas@ladybird.org>
Date: Thu, 24 Oct 2024 23:02:10 +0200
Subject: [PATCH] LibJS: Cache UTF-16 strings on the VM

We were already caching UTF-8 and byte strings, so let's add a cache
for UTF-16 strings as well. This is particularly profitable whenever we
run regular expressions, since the output of regex execution is a set of
UTF-16 strings.

Note that this is a weak cache like the other JS string caches, meaning
that strings are removed from the cache as they are garbage collected.

This avoids billions of PrimitiveString allocations across a run of WPT,
significantly reducing GC activity.
---
 .../LibJS/Runtime/PrimitiveString.cpp         | 10 +++++-
 .../Libraries/LibJS/Runtime/Utf16String.cpp   |  7 +++++
 .../Libraries/LibJS/Runtime/Utf16String.h     | 31 +++++++++++++++++++
 Userland/Libraries/LibJS/Runtime/VM.h         |  6 ++++
 4 files changed, 53 insertions(+), 1 deletion(-)

diff --git a/Userland/Libraries/LibJS/Runtime/PrimitiveString.cpp b/Userland/Libraries/LibJS/Runtime/PrimitiveString.cpp
index 68e7fd6e6b2..8eeb3c602e2 100644
--- a/Userland/Libraries/LibJS/Runtime/PrimitiveString.cpp
+++ b/Userland/Libraries/LibJS/Runtime/PrimitiveString.cpp
@@ -47,6 +47,8 @@ PrimitiveString::~PrimitiveString()
 {
     if (has_utf8_string())
         vm().string_cache().remove(*m_utf8_string);
+    if (has_utf16_string())
+        vm().utf16_string_cache().remove(*m_utf16_string);
     if (has_byte_string())
         vm().byte_string_cache().remove(*m_byte_string);
 }
@@ -167,7 +169,13 @@ NonnullGCPtr<PrimitiveString> PrimitiveString::create(VM& vm, Utf16String string
             return vm.single_ascii_character_string(static_cast<u8>(code_unit));
     }
 
-    return vm.heap().allocate_without_realm<PrimitiveString>(move(string));
+    auto& string_cache = vm.utf16_string_cache();
+    if (auto it = string_cache.find(string); it != string_cache.end())
+        return *it->value;
+
+    auto new_string = vm.heap().allocate_without_realm<PrimitiveString>(string);
+    string_cache.set(move(string), new_string);
+    return *new_string;
 }
 
 NonnullGCPtr<PrimitiveString> PrimitiveString::create(VM& vm, String string)
diff --git a/Userland/Libraries/LibJS/Runtime/Utf16String.cpp b/Userland/Libraries/LibJS/Runtime/Utf16String.cpp
index 295165e8f8c..2bd5a6c04ba 100644
--- a/Userland/Libraries/LibJS/Runtime/Utf16String.cpp
+++ b/Userland/Libraries/LibJS/Runtime/Utf16String.cpp
@@ -55,6 +55,13 @@ Utf16View Utf16StringImpl::view() const
     return Utf16View { m_string };
 }
 
+u32 Utf16StringImpl::compute_hash() const
+{
+    if (m_string.is_empty())
+        return 0;
+    return string_hash((char const*)m_string.data(), m_string.size() * sizeof(u16));
+}
+
 }
 
 Utf16String Utf16String::create()
diff --git a/Userland/Libraries/LibJS/Runtime/Utf16String.h b/Userland/Libraries/LibJS/Runtime/Utf16String.h
index 4e08afa4aa8..233a562ace6 100644
--- a/Userland/Libraries/LibJS/Runtime/Utf16String.h
+++ b/Userland/Libraries/LibJS/Runtime/Utf16String.h
@@ -29,10 +29,24 @@ public:
     Utf16Data const& string() const;
     Utf16View view() const;
 
+    [[nodiscard]] u32 hash() const
+    {
+        if (!m_has_hash) {
+            m_hash = compute_hash();
+            m_has_hash = true;
+        }
+        return m_hash;
+    }
+    [[nodiscard]] bool operator==(Utf16StringImpl const& other) const { return string() == other.string(); }
+
 private:
     Utf16StringImpl() = default;
     explicit Utf16StringImpl(Utf16Data string);
 
+    [[nodiscard]] u32 compute_hash() const;
+
+    mutable bool m_has_hash { false };
+    mutable u32 m_hash { 0 };
     Utf16Data m_string;
 };
 
@@ -57,6 +71,14 @@ public:
     size_t length_in_code_units() const;
     bool is_empty() const;
 
+    [[nodiscard]] u32 hash() const { return m_string->hash(); }
+    [[nodiscard]] bool operator==(Utf16String const& other) const
+    {
+        if (m_string == other.m_string)
+            return true;
+        return *m_string == *other.m_string;
+    }
+
 private:
     explicit Utf16String(NonnullRefPtr<Detail::Utf16StringImpl>);
 
@@ -64,3 +86,12 @@ private:
 };
 
 }
+
+namespace AK {
+
+template<>
+struct Traits<JS::Utf16String> : public DefaultTraits<JS::Utf16String> {
+    static unsigned hash(JS::Utf16String const& s) { return s.hash(); }
+};
+
+}
diff --git a/Userland/Libraries/LibJS/Runtime/VM.h b/Userland/Libraries/LibJS/Runtime/VM.h
index d39085111fc..ee6a8a4b73d 100644
--- a/Userland/Libraries/LibJS/Runtime/VM.h
+++ b/Userland/Libraries/LibJS/Runtime/VM.h
@@ -75,6 +75,11 @@ public:
         return m_byte_string_cache;
     }
 
+    HashMap<Utf16String, GCPtr<PrimitiveString>>& utf16_string_cache()
+    {
+        return m_utf16_string_cache;
+    }
+
     PrimitiveString& empty_string() { return *m_empty_string; }
 
     PrimitiveString& single_ascii_character_string(u8 character)
@@ -298,6 +303,7 @@ private:
 
     HashMap<String, GCPtr<PrimitiveString>> m_string_cache;
     HashMap<ByteString, GCPtr<PrimitiveString>> m_byte_string_cache;
+    HashMap<Utf16String, GCPtr<PrimitiveString>> m_utf16_string_cache;
 
     Heap m_heap;