[dynarmic] remove decode matcher function handlers using std::function<>, use raw function pointers (#3920)

issues:
- std::function<> is used, which is famously bad
- storage of tehse in tables makes big fucking tables for no good reason
- lets just store a normal pointer and stuff! :)

this pr attempts to address that

Signed-off-by: lizzie <lizzie@eden-emu.dev>
Reviewed-on: https://git.eden-emu.dev/eden-emu/eden/pulls/3920
Reviewed-by: CamilleLaVey <camillelavey99@gmail.com>

15 files changed, 187 insertions, 276 deletions

diff --git a/src/core/arm/nce/interpreter_visitor.cpp b/src/core/arm/nce/interpreter_visitor.cpp
index 077a696cc8..ec2b155fe4 100644
--- a/src/core/arm/nce/interpreter_visitor.cpp
+++ b/src/core/arm/nce/interpreter_visitor.cpp
@@ -761,8 +761,7 @@ bool InterpreterVisitor::LDR_reg_fpsimd(Imm<2> size, Imm<1> opc_1, Reg Rm, Imm<3
     return this->SIMDOffset(scale, shift, opc_0, Rm, option, Rn, Vt);
 }
 
-std::optional<u64> MatchAndExecuteOneInstruction(Core::Memory::Memory& memory, mcontext_t* context,
-                                                 fpsimd_context* fpsimd_context) {
+std::optional<u64> MatchAndExecuteOneInstruction(Core::Memory::Memory& memory, mcontext_t* context, fpsimd_context* fpsimd_context) {
     std::span<u64, 31> regs(reinterpret_cast<u64*>(context->regs), 31);
     std::span<u128, 32> vregs(reinterpret_cast<u128*>(fpsimd_context->vregs), 32);
     u64& sp = *reinterpret_cast<u64*>(&context->sp);
@@ -772,9 +771,9 @@ std::optional<u64> MatchAndExecuteOneInstruction(Core::Memory::Memory& memory, m
     u32 instruction = memory.Read32(pc);
     bool was_executed = false;
 
-    auto decoder = Dynarmic::A64::Decode<VisitorBase>(instruction);
+    auto decoder = Dynarmic::A64::Decode<VisitorBase, bool>(visitor, instruction);
     if (decoder) {
-        was_executed = decoder->get().call(visitor, instruction);
+        was_executed = *decoder;
     } else {
         was_executed = false;
     }
diff --git a/src/dynarmic/src/dynarmic/frontend/A32/decoder/arm.h b/src/dynarmic/src/dynarmic/frontend/A32/decoder/arm.h
index d1cd9a923d..a9309ef5a8 100644
--- a/src/dynarmic/src/dynarmic/frontend/A32/decoder/arm.h
+++ b/src/dynarmic/src/dynarmic/frontend/A32/decoder/arm.h
@@ -26,48 +26,41 @@ namespace Dynarmic::A32 {
 template<typename Visitor>
 using ArmMatcher = Decoder::Matcher<Visitor, u32>;
 
-template<typename Visitor>
-using ArmDecodeTable = std::array<std::vector<ArmMatcher<Visitor>>, 0x1000>;
-
-namespace detail {
-inline size_t ToFastLookupIndexArm(u32 instruction) noexcept {
-    return ((instruction >> 4) & 0x00F) | ((instruction >> 16) & 0xFF0);
-}
-}  // namespace detail
-
-template<typename V>
-static ArmDecodeTable<V> GetArmDecodeTable() noexcept {
-    ArmDecodeTable<V> table{};
-    for (size_t i = 0; i < table.size(); ++i) {
-        // PLEASE HEAP ELLIDE
-        for (auto const& e : std::vector<ArmMatcher<V>>{
-#define INST(fn, name, bitstring) DYNARMIC_DECODER_GET_MATCHER(ArmMatcher, fn, name, Decoder::detail::StringToArray<32>(bitstring)),
+template<typename V, typename ReturnType>
+static std::optional<ReturnType> DecodeArm(V& visitor, u32 instruction) noexcept {
+    auto const make_fast_index = [](u32 a) {
+        return ((a >> 4) & 0x00F) | ((a >> 16) & 0xFF0);
+    };
+    struct Handler {
+        bool (*fn)(V&, u32);
+        u32 mask;
+        u32 expect;
+    };
+    alignas(64) static const std::array<std::vector<Handler>, 0x1000> table = [&] {
+        std::array<std::vector<Handler>, 0x1000> t{};
+        for (size_t i = 0; i < t.size(); ++i) {
+#define INST(fn, name, bitstring) \
+    do { \
+        auto const [mask, expect] = DYNARMIC_DECODER_GET_MATCHER(ArmMatcher, fn, name, Decoder::detail::StringToArray<32>(bitstring)); \
+        if ((i & make_fast_index(mask)) == make_fast_index(expect)) { \
+            t[i].emplace_back([](V& visitor, u32 instruction) -> bool { \
+                return DYNARMIC_DECODER_GET_MATCHER_FUNCTION(ArmMatcher, fn, name, Decoder::detail::StringToArray<32>(bitstring)); \
+            }, mask, expect); \
+        } \
+    } while (0);
 #include "./arm.inc"
 #undef INST
-        }) {
-            auto const expect = detail::ToFastLookupIndexArm(e.GetExpected());
-            auto const mask = detail::ToFastLookupIndexArm(e.GetMask());
-            if ((i & mask) == expect) {
-                table[i].push_back(e);
-            }
         }
-    }
-    return table;
-}
-
-template<typename V>
-static std::optional<std::reference_wrapper<const ArmMatcher<V>>> DecodeArm(u32 instruction) noexcept {
-    alignas(64) static const auto table = GetArmDecodeTable<V>();
-    const auto matches_instruction = [instruction](const auto& matcher) {
-        return matcher.Matches(instruction);
-    };
-    const auto& subtable = table[detail::ToFastLookupIndexArm(instruction)];
-    auto iter = std::find_if(subtable.begin(), subtable.end(), matches_instruction);
-    return iter != subtable.end() ? std::optional<std::reference_wrapper<const ArmMatcher<V>>>(*iter) : std::nullopt;
+        return t;
+    }();
+    for (auto const& e : table[make_fast_index(instruction)])
+        if ((instruction & e.mask) == e.expect)
+            return e.fn(visitor, instruction);
+    return std::nullopt;
 }
 
 template<typename V>
-static std::optional<std::string_view> GetNameARM(u32 inst) noexcept {
+static std::optional<std::string_view> GetNameArm(u32 inst) noexcept {
     std::vector<std::pair<std::string_view, ArmMatcher<V>>> list = {
 #define INST(fn, name, bitstring) { name, DYNARMIC_DECODER_GET_MATCHER(ArmMatcher, fn, name, Decoder::detail::StringToArray<32>(bitstring)) },
 #include "./arm.inc"
diff --git a/src/dynarmic/src/dynarmic/frontend/A32/decoder/asimd.h b/src/dynarmic/src/dynarmic/frontend/A32/decoder/asimd.h
index 5b0336e59f..4feae6f478 100644
--- a/src/dynarmic/src/dynarmic/frontend/A32/decoder/asimd.h
+++ b/src/dynarmic/src/dynarmic/frontend/A32/decoder/asimd.h
@@ -26,17 +26,16 @@ namespace Dynarmic::A32 {
 template<typename Visitor>
 using ASIMDMatcher = Decoder::Matcher<Visitor, u32>;
 
-template<typename V>
-static std::optional<std::reference_wrapper<const ASIMDMatcher<V>>> DecodeASIMD(u32 instruction) noexcept {
-    alignas(64) static const auto table = std::array{
-#define INST(fn, name, bitstring) DYNARMIC_DECODER_GET_MATCHER(ASIMDMatcher, fn, name, Decoder::detail::StringToArray<32>(bitstring)),
+template<typename V, typename ReturnType>
+static std::optional<ReturnType> DecodeASIMD(V& visitor, u32 instruction) noexcept {
+#define INST(fn, name, bitstring) \
+    do { \
+        auto const [mask, expect] = DYNARMIC_DECODER_GET_MATCHER(ASIMDMatcher, fn, name, Decoder::detail::StringToArray<32>(bitstring)); \
+        if ((instruction & mask) == expect) return DYNARMIC_DECODER_GET_MATCHER_FUNCTION(ASIMDMatcher, fn, name, Decoder::detail::StringToArray<32>(bitstring)); \
+    } while (0);
 #include "./asimd.inc"
 #undef INST
-    };
-    auto iter = std::find_if(table.begin(), table.end(), [instruction](const auto& matcher) {
-        return matcher.Matches(instruction);
-    });
-    return iter != table.end() ? std::optional<std::reference_wrapper<const ASIMDMatcher<V>>>(*iter) : std::nullopt;
+    return std::nullopt;
 }
 
 template<typename V>
diff --git a/src/dynarmic/src/dynarmic/frontend/A32/decoder/thumb16.h b/src/dynarmic/src/dynarmic/frontend/A32/decoder/thumb16.h
index 590878a6de..465159a00d 100644
--- a/src/dynarmic/src/dynarmic/frontend/A32/decoder/thumb16.h
+++ b/src/dynarmic/src/dynarmic/frontend/A32/decoder/thumb16.h
@@ -23,17 +23,16 @@ namespace Dynarmic::A32 {
 template<typename Visitor>
 using Thumb16Matcher = Decoder::Matcher<Visitor, u16>;
 
-template<typename V>
-static std::optional<std::reference_wrapper<const Thumb16Matcher<V>>> DecodeThumb16(u16 instruction) {
-    alignas(64) static const auto table = std::array{
-#define INST(fn, name, bitstring) DYNARMIC_DECODER_GET_MATCHER(Thumb16Matcher, fn, name, Decoder::detail::StringToArray<16>(bitstring)),
+template<typename V, typename ReturnType>
+static std::optional<ReturnType> DecodeThumb16(V& visitor, u16 instruction) {
+#define INST(fn, name, bitstring) \
+    do { \
+        auto const [mask, expect] = DYNARMIC_DECODER_GET_MATCHER(Thumb16Matcher, fn, name, Decoder::detail::StringToArray<16>(bitstring)); \
+        if ((instruction & mask) == expect) return DYNARMIC_DECODER_GET_MATCHER_FUNCTION(Thumb16Matcher, fn, name, Decoder::detail::StringToArray<16>(bitstring)); \
+    } while (0);
 #include "./thumb16.inc"
 #undef INST
-    };
-    auto iter = std::find_if(table.begin(), table.end(), [instruction](const auto& matcher) {
-        return matcher.Matches(instruction);
-    });
-    return iter != table.end() ? std::optional<std::reference_wrapper<const Thumb16Matcher<V>>>(*iter) : std::nullopt;
+    return std::nullopt;
 }
 
 template<typename V>
diff --git a/src/dynarmic/src/dynarmic/frontend/A32/decoder/thumb32.h b/src/dynarmic/src/dynarmic/frontend/A32/decoder/thumb32.h
index 6a09c66afb..b3588c7a1d 100644
--- a/src/dynarmic/src/dynarmic/frontend/A32/decoder/thumb32.h
+++ b/src/dynarmic/src/dynarmic/frontend/A32/decoder/thumb32.h
@@ -22,17 +22,16 @@ namespace Dynarmic::A32 {
 template<typename Visitor>
 using Thumb32Matcher = Decoder::Matcher<Visitor, u32>;
 
-template<typename V>
-static std::optional<std::reference_wrapper<const Thumb32Matcher<V>>> DecodeThumb32(u32 instruction) {
-    alignas(64) static const auto table = std::array{
-#define INST(fn, name, bitstring) DYNARMIC_DECODER_GET_MATCHER(Thumb32Matcher, fn, name, Decoder::detail::StringToArray<32>(bitstring)),
+template<typename V, typename ReturnType>
+static std::optional<ReturnType> DecodeThumb32(V& visitor, u32 instruction) {
+#define INST(fn, name, bitstring) \
+    do { \
+        auto const [mask, expect] = DYNARMIC_DECODER_GET_MATCHER(Thumb32Matcher, fn, name, Decoder::detail::StringToArray<32>(bitstring)); \
+        if ((instruction & mask) == expect) return DYNARMIC_DECODER_GET_MATCHER_FUNCTION(Thumb32Matcher, fn, name, Decoder::detail::StringToArray<32>(bitstring)); \
+    } while (0);
 #include "./thumb32.inc"
 #undef INST
-    };
-    auto iter = std::find_if(table.begin(), table.end(), [instruction](const auto& matcher) {
-        return matcher.Matches(instruction);
-    });
-    return iter != table.end() ? std::optional<std::reference_wrapper<const Thumb32Matcher<V>>>(*iter) : std::nullopt;
+    return std::nullopt;
 }
 
 template<typename V>
diff --git a/src/dynarmic/src/dynarmic/frontend/A32/decoder/vfp.h b/src/dynarmic/src/dynarmic/frontend/A32/decoder/vfp.h
index e02cc03a6b..8703aa24b4 100644
--- a/src/dynarmic/src/dynarmic/frontend/A32/decoder/vfp.h
+++ b/src/dynarmic/src/dynarmic/frontend/A32/decoder/vfp.h
@@ -23,32 +23,18 @@ namespace Dynarmic::A32 {
 template<typename Visitor>
 using VFPMatcher = Decoder::Matcher<Visitor, u32>;
 
-template<typename V>
-static std::optional<std::reference_wrapper<const VFPMatcher<V>>> DecodeVFP(u32 instruction) {
-    using Table = std::vector<VFPMatcher<V>>;
-    alignas(64) static const struct Tables {
-        Table unconditional;
-        Table conditional;
-    } tables = []() {
-        Table list = {
-#define INST(fn, name, bitstring) DYNARMIC_DECODER_GET_MATCHER(VFPMatcher, fn, name, Decoder::detail::StringToArray<32>(bitstring)),
+template<typename V, typename ReturnType>
+static std::optional<ReturnType> DecodeVFP(V& visitor, u32 instruction) {
+    bool const i_uncond = (instruction & 0xF0000000) == 0xF0000000;
+#define INST(fn, name, bitstring) \
+    do { \
+        auto const [mask, expect] = DYNARMIC_DECODER_GET_MATCHER(VFPMatcher, fn, name, Decoder::detail::StringToArray<32>(bitstring)); \
+        bool const m_uncond = (mask & 0xF0000000) == 0xF0000000; \
+        if ((instruction & mask) == expect && m_uncond == i_uncond) return DYNARMIC_DECODER_GET_MATCHER_FUNCTION(VFPMatcher, fn, name, Decoder::detail::StringToArray<32>(bitstring)); \
+    } while (0);
 #include "./vfp.inc"
 #undef INST
-        };
-        auto const it = std::stable_partition(list.begin(), list.end(), [&](const auto& matcher) {
-            return (matcher.GetMask() & 0xF0000000) == 0xF0000000;
-        });
-        return Tables{
-            Table{list.begin(), it},
-            Table{it, list.end()},
-        };
-    }();
-    const bool is_unconditional = (instruction & 0xF0000000) == 0xF0000000;
-    const Table& table = is_unconditional ? tables.unconditional : tables.conditional;
-    auto iter = std::find_if(table.begin(), table.end(), [instruction](const auto& matcher) {
-        return matcher.Matches(instruction);
-    });
-    return iter != table.end() ? std::optional<std::reference_wrapper<const VFPMatcher<V>>>(*iter) : std::nullopt;
+    return std::nullopt;
 }
 
 template<typename V>
diff --git a/src/dynarmic/src/dynarmic/frontend/A32/translate/translate_arm.cpp b/src/dynarmic/src/dynarmic/frontend/A32/translate/translate_arm.cpp
index 3fd475d074..6c89a4af36 100644
--- a/src/dynarmic/src/dynarmic/frontend/A32/translate/translate_arm.cpp
+++ b/src/dynarmic/src/dynarmic/frontend/A32/translate/translate_arm.cpp
@@ -41,12 +41,12 @@ void TranslateArm(IR::Block& block, LocationDescriptor descriptor, TranslateCall
             tcb->PreCodeTranslationHook(false, arm_pc, visitor.ir);
             ticks_for_instruction = tcb->GetTicksForCode(false, arm_pc, *arm_instruction);
 
-            if (const auto vfp_decoder = DecodeVFP<TranslatorVisitor>(*arm_instruction)) {
-                should_continue = vfp_decoder->get().call(visitor, *arm_instruction);
-            } else if (const auto asimd_decoder = DecodeASIMD<TranslatorVisitor>(*arm_instruction)) {
-                should_continue = asimd_decoder->get().call(visitor, *arm_instruction);
-            } else if (const auto decoder = DecodeArm<TranslatorVisitor>(*arm_instruction)) {
-                should_continue = decoder->get().call(visitor, *arm_instruction);
+            if (const auto vfp_decoder = DecodeVFP<TranslatorVisitor, bool>(visitor, *arm_instruction)) {
+                should_continue = *vfp_decoder;
+            } else if (const auto asimd_decoder = DecodeASIMD<TranslatorVisitor, bool>(visitor, *arm_instruction)) {
+                should_continue = *asimd_decoder;
+            } else if (const auto decoder = DecodeArm<TranslatorVisitor, bool>(visitor, *arm_instruction)) {
+                should_continue = *decoder;
             } else {
                 should_continue = visitor.arm_UDF();
             }
@@ -88,12 +88,12 @@ bool TranslateSingleArmInstruction(IR::Block& block, LocationDescriptor descript
 
     const u64 ticks_for_instruction = 1;
 
-    if (const auto vfp_decoder = DecodeVFP<TranslatorVisitor>(arm_instruction)) {
-        should_continue = vfp_decoder->get().call(visitor, arm_instruction);
-    } else if (const auto asimd_decoder = DecodeASIMD<TranslatorVisitor>(arm_instruction)) {
-        should_continue = asimd_decoder->get().call(visitor, arm_instruction);
-    } else if (const auto decoder = DecodeArm<TranslatorVisitor>(arm_instruction)) {
-        should_continue = decoder->get().call(visitor, arm_instruction);
+    if (const auto vfp_decoder = DecodeVFP<TranslatorVisitor, bool>(visitor, arm_instruction)) {
+        should_continue = *vfp_decoder;
+    } else if (const auto asimd_decoder = DecodeASIMD<TranslatorVisitor, bool>(visitor, arm_instruction)) {
+        should_continue = *asimd_decoder;
+    } else if (const auto decoder = DecodeArm<TranslatorVisitor, bool>(visitor, arm_instruction)) {
+        should_continue = *decoder;
     } else {
         should_continue = visitor.arm_UDF();
     }
diff --git a/src/dynarmic/src/dynarmic/frontend/A32/translate/translate_thumb.cpp b/src/dynarmic/src/dynarmic/frontend/A32/translate/translate_thumb.cpp
index 332860a95c..2a2fbea63f 100644
--- a/src/dynarmic/src/dynarmic/frontend/A32/translate/translate_thumb.cpp
+++ b/src/dynarmic/src/dynarmic/frontend/A32/translate/translate_thumb.cpp
@@ -126,24 +126,24 @@ void TranslateThumb(IR::Block& block, LocationDescriptor descriptor, TranslateCa
 
             if (IsUnconditionalInstruction(is_thumb_16, thumb_instruction) || visitor.ThumbConditionPassed()) {
                 if (is_thumb_16) {
-                    if (const auto decoder = DecodeThumb16<TranslatorVisitor>(static_cast<u16>(thumb_instruction))) {
-                        should_continue = decoder->get().call(visitor, static_cast<u16>(thumb_instruction));
+                    if (const auto decoder = DecodeThumb16<TranslatorVisitor, bool>(visitor, u16(thumb_instruction))) {
+                        should_continue = *decoder;
                     } else {
                         should_continue = visitor.thumb16_UDF();
                     }
                 } else {
                     if (MaybeVFPOrASIMDInstruction(thumb_instruction)) {
-                        if (const auto vfp_decoder = DecodeVFP<TranslatorVisitor>(thumb_instruction)) {
-                            should_continue = vfp_decoder->get().call(visitor, thumb_instruction);
-                        } else if (const auto asimd_decoder = DecodeASIMD<TranslatorVisitor>(ConvertASIMDInstruction(thumb_instruction))) {
-                            should_continue = asimd_decoder->get().call(visitor, ConvertASIMDInstruction(thumb_instruction));
-                        } else if (const auto decoder = DecodeThumb32<TranslatorVisitor>(thumb_instruction)) {
-                            should_continue = decoder->get().call(visitor, thumb_instruction);
+                        if (const auto vfp_decoder = DecodeVFP<TranslatorVisitor, bool>(visitor, thumb_instruction)) {
+                            should_continue = *vfp_decoder;
+                        } else if (const auto asimd_decoder = DecodeASIMD<TranslatorVisitor, bool>(visitor, ConvertASIMDInstruction(thumb_instruction))) {
+                            should_continue = *asimd_decoder;
+                        } else if (const auto decoder = DecodeThumb32<TranslatorVisitor, bool>(visitor, thumb_instruction)) {
+                            should_continue = *decoder;
                         } else {
                             should_continue = visitor.thumb32_UDF();
                         }
-                    } else if (const auto decoder = DecodeThumb32<TranslatorVisitor>(thumb_instruction)) {
-                        should_continue = decoder->get().call(visitor, thumb_instruction);
+                    } else if (const auto decoder = DecodeThumb32<TranslatorVisitor, bool>(visitor, thumb_instruction)) {
+                        should_continue = *decoder;
                     } else {
                         should_continue = visitor.thumb32_UDF();
                     }
@@ -187,25 +187,25 @@ bool TranslateSingleThumbInstruction(IR::Block& block, LocationDescriptor descri
     const u64 ticks_for_instruction = 1;
 
     if (is_thumb_16) {
-        if (const auto decoder = DecodeThumb16<TranslatorVisitor>(static_cast<u16>(thumb_instruction))) {
-            should_continue = decoder->get().call(visitor, static_cast<u16>(thumb_instruction));
+        if (const auto decoder = DecodeThumb16<TranslatorVisitor, bool>(visitor, u16(thumb_instruction))) {
+            should_continue = *decoder;
         } else {
             should_continue = visitor.thumb16_UDF();
         }
     } else {
         thumb_instruction = mcl::bit::swap_halves_32(thumb_instruction);
         if (MaybeVFPOrASIMDInstruction(thumb_instruction)) {
-            if (const auto vfp_decoder = DecodeVFP<TranslatorVisitor>(thumb_instruction)) {
-                should_continue = vfp_decoder->get().call(visitor, thumb_instruction);
-            } else if (const auto asimd_decoder = DecodeASIMD<TranslatorVisitor>(ConvertASIMDInstruction(thumb_instruction))) {
-                should_continue = asimd_decoder->get().call(visitor, ConvertASIMDInstruction(thumb_instruction));
-            } else if (const auto decoder = DecodeThumb32<TranslatorVisitor>(thumb_instruction)) {
-                should_continue = decoder->get().call(visitor, thumb_instruction);
+            if (const auto vfp_decoder = DecodeVFP<TranslatorVisitor, bool>(visitor, thumb_instruction)) {
+                should_continue = *vfp_decoder;
+            } else if (const auto asimd_decoder = DecodeASIMD<TranslatorVisitor, bool>(visitor, ConvertASIMDInstruction(thumb_instruction))) {
+                should_continue = *asimd_decoder;
+            } else if (const auto decoder = DecodeThumb32<TranslatorVisitor, bool>(visitor, thumb_instruction)) {
+                should_continue = *decoder;
             } else {
                 should_continue = visitor.thumb32_UDF();
             }
-        } else if (const auto decoder = DecodeThumb32<TranslatorVisitor>(thumb_instruction)) {
-            should_continue = decoder->get().call(visitor, thumb_instruction);
+        } else if (const auto decoder = DecodeThumb32<TranslatorVisitor, bool>(visitor, thumb_instruction)) {
+            should_continue = *decoder;
         } else {
             should_continue = visitor.thumb32_UDF();
         }
diff --git a/src/dynarmic/src/dynarmic/frontend/A64/decoder/a64.h b/src/dynarmic/src/dynarmic/frontend/A64/decoder/a64.h
index ccf890cbcd..f0b7baba7b 100644
--- a/src/dynarmic/src/dynarmic/frontend/A64/decoder/a64.h
+++ b/src/dynarmic/src/dynarmic/frontend/A64/decoder/a64.h
@@ -26,45 +26,37 @@ namespace Dynarmic::A64 {
 template<typename Visitor>
 using Matcher = Decoder::Matcher<Visitor, u32>;
 
-template<typename Visitor>
-using DecodeTable = std::array<std::vector<Matcher<Visitor>>, 0x1000>;
-
-namespace detail {
-inline size_t ToFastLookupIndex(u32 instruction) {
-    return ((instruction >> 10) & 0x00F) | ((instruction >> 18) & 0xFF0);
-}
-}  // namespace detail
-
-template<typename V>
-inline DecodeTable<V> GetDecodeTable() {
-    DecodeTable<V> table{};
-    for (size_t i = 0; i < table.size(); ++i) {
-        // PLEASE HEAP ELLIDE
-        for (auto const& e : std::vector<Matcher<V>>{
-#define INST(fn, name, bitstring) DYNARMIC_DECODER_GET_MATCHER(Matcher, fn, name, Decoder::detail::StringToArray<32>(bitstring)),
+template<typename V, typename ReturnType>
+static std::optional<ReturnType> Decode(V& visitor, u32 instruction) noexcept {
+    auto const make_fast_index = [](u32 a) {
+        return ((a >> 10) & 0x00F) | ((a >> 18) & 0xFF0);
+    };
+    struct Handler {
+        bool (*fn)(V&, u32);
+        u32 mask;
+        u32 expect;
+    };
+    alignas(64) static const std::array<std::vector<Handler>, 0x1000> table = [&] {
+        std::array<std::vector<Handler>, 0x1000> t{};
+        for (size_t i = 0; i < t.size(); ++i) {
+#define INST(fn, name, bitstring) \
+    do { \
+        auto const [mask, expect] = DYNARMIC_DECODER_GET_MATCHER(Matcher, fn, name, Decoder::detail::StringToArray<32>(bitstring)); \
+        if ((i & make_fast_index(mask)) == make_fast_index(expect)) { \
+            t[i].emplace_back([](V& visitor, u32 instruction) -> bool { \
+                return DYNARMIC_DECODER_GET_MATCHER_FUNCTION(Matcher, fn, name, Decoder::detail::StringToArray<32>(bitstring)); \
+            }, mask, expect); \
+        } \
+    } while (0);
 #include "./a64.inc"
 #undef INST
-        }) {
-            const auto expect = detail::ToFastLookupIndex(e.GetExpected());
-            const auto mask = detail::ToFastLookupIndex(e.GetMask());
-            if ((i & mask) == expect)
-                table[i].push_back(e);
         }
-    }
-    return table;
-}
-
-/// In practice it must always suceed, otherwise something else unrelated would have gone awry
-template<typename V>
-inline std::optional<std::reference_wrapper<const Matcher<V>>> Decode(u32 instruction) {
-    alignas(64) static const auto table = GetDecodeTable<V>();
-    const auto& subtable = table[detail::ToFastLookupIndex(instruction)];
-    auto iter = std::find_if(subtable.begin(), subtable.end(), [instruction](const auto& matcher) {
-        return matcher.Matches(instruction);
-    });
-    return iter != subtable.end()
-        ? std::optional{ std::reference_wrapper<const Matcher<V>>(*iter) }
-        : std::nullopt;
+        return t;
+    }();
+    for (auto const& e : table[make_fast_index(instruction)])
+        if ((instruction & e.mask) == e.expect)
+            return e.fn(visitor, instruction);
+    return std::nullopt;
 }
 
 template<typename V>
diff --git a/src/dynarmic/src/dynarmic/frontend/A64/translate/a64_translate.cpp b/src/dynarmic/src/dynarmic/frontend/A64/translate/a64_translate.cpp
index 4afce6bd29..7840c63d79 100644
--- a/src/dynarmic/src/dynarmic/frontend/A64/translate/a64_translate.cpp
+++ b/src/dynarmic/src/dynarmic/frontend/A64/translate/a64_translate.cpp
@@ -24,9 +24,8 @@ void Translate(IR::Block& block, LocationDescriptor descriptor, MemoryReadCodeFu
     do {
         const u64 pc = visitor.ir.current_location->PC();
         if (const auto instruction = memory_read_code(pc)) {
-            auto decoder = Decode<TranslatorVisitor>(*instruction);
-            if (decoder) {
-                should_continue = decoder->get().call(visitor, *instruction);
+            if (auto decoder = Decode<TranslatorVisitor, bool>(visitor, *instruction)) {
+                should_continue = *decoder;
             } else {
                 should_continue = visitor.RaiseException(Exception::UnallocatedEncoding);
             }
@@ -47,13 +46,9 @@ void Translate(IR::Block& block, LocationDescriptor descriptor, MemoryReadCodeFu
 bool TranslateSingleInstruction(IR::Block& block, LocationDescriptor descriptor, u32 instruction) {
     TranslatorVisitor visitor{block, descriptor, {}};
 
-    bool should_continue = true;
-    auto const decoder = Decode<TranslatorVisitor>(instruction);
-    if (decoder) {
-        should_continue = decoder->get().call(visitor, instruction);
-    } else {
-        should_continue = false;
-    }
+    bool should_continue = false;
+    if (auto const decoder = Decode<TranslatorVisitor, bool>(visitor, instruction))
+        should_continue = *decoder;
 
     visitor.ir.current_location = visitor.ir.current_location->AdvancePC(4);
     block.CycleCount()++;
diff --git a/src/dynarmic/src/dynarmic/frontend/decoder/decoder_detail.h b/src/dynarmic/src/dynarmic/frontend/decoder/decoder_detail.h
index c22e585c26..e383fb0262 100644
--- a/src/dynarmic/src/dynarmic/frontend/decoder/decoder_detail.h
+++ b/src/dynarmic/src/dynarmic/frontend/decoder/decoder_detail.h
@@ -8,31 +8,25 @@
 
 #pragma once
 
-#include <algorithm>
 #include <array>
 #include <tuple>
-
+#include <utility>
 #include "common/assert.h"
 #include "dynarmic/mcl/bit.hpp"
-#include "dynarmic/mcl/function_info.hpp"
 
 namespace Dynarmic::Decoder {
 namespace detail {
 
 template<size_t N>
-inline consteval std::array<char, N> StringToArray(const char (&str)[N + 1]) {
-    std::array<char, N> result{};
-    for (size_t i = 0; i < N; i++) {
-        result[i] = str[i];
-    }
-    return result;
+inline consteval std::array<char, N> StringToArray(const char (&s)[N + 1]) {
+    std::array<char, N> r{};
+    for (size_t i = 0; i < N; i++)
+        r[i] = s[i];
+    return r;
 }
 
-/**
- * Helper functions for the decoders.
- *
- * @tparam MatcherT The type of the Matcher to use.
- */
+/// @brief Helper functions for the decoders.
+/// @tparam MatcherT The type of the Matcher to use.
 template<class MatcherT>
 struct detail {
     using opcode_type = typename MatcherT::opcode_type;
@@ -40,17 +34,15 @@ struct detail {
 
     static constexpr size_t opcode_bitsize = mcl::bitsizeof<opcode_type>;
 
-    /**
-     * Generates the mask and the expected value after masking from a given bitstring.
-     * A '0' in a bitstring indicates that a zero must be present at that bit position.
-     * A '1' in a bitstring indicates that a one must be present at that bit position.
-     */
+    /// @brief Generates the mask and the expected value after masking from a given bitstring.
+    /// A '0' in a bitstring indicates that a zero must be present at that bit position.
+    /// A '1' in a bitstring indicates that a one must be present at that bit position.
 #ifdef __clang__
     static constexpr auto GetMaskAndExpect(std::array<char, opcode_bitsize> bitstring) {
 #else
     static consteval auto GetMaskAndExpect(std::array<char, opcode_bitsize> bitstring) {
 #endif
-        const auto one = static_cast<opcode_type>(1);
+        const auto one = opcode_type(1);
         opcode_type mask = 0, expect = 0;
         for (size_t i = 0; i < opcode_bitsize; i++) {
             const size_t bit_position = opcode_bitsize - i - 1;
@@ -101,7 +93,7 @@ struct detail {
         }
 #if !defined(DYNARMIC_IGNORE_ASSERTS) && !defined(__ANDROID__)
         // Avoids a MSVC ICE, and avoids Android NDK issue.
-        ASSERT(std::all_of(masks.begin(), masks.end(), [](auto m) { return m != 0; }));
+        DEBUG_ASSERT(std::all_of(masks.begin(), masks.end(), [](auto m) { return m != 0; }));
 #endif
         return std::make_tuple(masks, shifts);
     }
@@ -109,65 +101,32 @@ struct detail {
     /// @brief This struct's Make member function generates a lambda which decodes an instruction
     /// based on the provided arg_masks and arg_shifts. The Visitor member function to call is
     /// provided as a template argument.
-    template<typename FnT>
-    struct VisitorCaller;
-
 #ifdef _MSC_VER
 #    pragma warning(push)
 #    pragma warning(disable : 4800)  // forcing value to bool 'true' or 'false' (performance warning)
 #endif
-    template<typename V, typename... Args, typename ReturnType>
-    struct VisitorCaller<ReturnType (V::*)(Args...)> {
-        template<size_t... iota>
-        static constexpr auto Make(std::integer_sequence<size_t, iota...>,
-                         ReturnType (V::*const fn)(Args...),
-                         const std::array<opcode_type, sizeof...(iota)> arg_masks,
-                         const std::array<size_t, sizeof...(iota)> arg_shifts) {
-            static_assert(std::is_same_v<visitor_type, V>, "Member function is not from Matcher's Visitor");
-            return [fn, arg_masks, arg_shifts](V& v, opcode_type instruction) {
-                (void)instruction;
-                (void)arg_masks;
-                (void)arg_shifts;
-                return (v.*fn)(Args((instruction & arg_masks[iota]) >> arg_shifts[iota])...);
-            };
-        }
-    };
-
-    template<typename V, typename... Args, typename ReturnType>
-    struct VisitorCaller<ReturnType (V::*)(Args...) const> {
-        template<size_t... iota>
-        static constexpr auto Make(std::integer_sequence<size_t, iota...>,
-                         ReturnType (V::*const fn)(Args...) const,
-                         const std::array<opcode_type, sizeof...(iota)> arg_masks,
-                         const std::array<size_t, sizeof...(iota)> arg_shifts) {
-            static_assert(std::is_same_v<visitor_type, const V>, "Member function is not from Matcher's Visitor");
-            return [fn, arg_masks, arg_shifts](const V& v, opcode_type instruction) {
-                (void)instruction;
-                (void)arg_masks;
-                (void)arg_shifts;
-                return (v.*fn)(Args((instruction & arg_masks[iota]) >> arg_shifts[iota])...);
-            };
+    template<typename V, typename ReturnType, typename... Args>
+    struct VisitorCaller {
+        template<std::size_t... iota>
+        static inline constexpr auto Invoke(std::index_sequence<iota...>, V& visitor, [[maybe_unused]] opcode_type instruction, ReturnType (V::*const fn)(Args...), [[maybe_unused]] const std::array<opcode_type, sizeof...(Args)> arg_masks, [[maybe_unused]] const std::array<size_t, sizeof...(Args)> arg_shifts) {
+            return (visitor.*fn)(Args((instruction & arg_masks[iota]) >> arg_shifts[iota])...);
         }
     };
 #ifdef _MSC_VER
 #    pragma warning(pop)
 #endif
 
-    /// @brief Creates a matcher that can match and parse instructions based on bitstring.
-    /// See also: GetMaskAndExpect and GetArgInfo for format of bitstring.
-    template<auto bitstring, typename F>
-    static constexpr auto GetMatcher(F fn) {
-        constexpr size_t args_count = mcl::parameter_count_v<F>;
-        constexpr auto mask = std::get<0>(GetMaskAndExpect(bitstring));
-        constexpr auto expect = std::get<1>(GetMaskAndExpect(bitstring));
-        constexpr auto arg_masks = std::get<0>(GetArgInfo<args_count>(bitstring));
-        constexpr auto arg_shifts = std::get<1>(GetArgInfo<args_count>(bitstring));
-        const auto proxy_fn = VisitorCaller<F>::Make(std::make_index_sequence<args_count>(), fn, arg_masks, arg_shifts);
-        return MatcherT(mask, expect, proxy_fn);
+    template<auto bitstring, typename V, typename ReturnType, typename... Args>
+    static inline constexpr auto GetMatcherFunction(V& visitor, [[maybe_unused]] opcode_type instruction, ReturnType (V::*const fn)(Args...)) {
+        constexpr auto arg_masks = std::get<0>(GetArgInfo<sizeof...(Args)>(bitstring));
+        constexpr auto arg_shifts = std::get<1>(GetArgInfo<sizeof...(Args)>(bitstring));
+        return VisitorCaller<V, ReturnType, Args...>::Invoke(std::index_sequence_for<Args...>(), visitor, instruction, fn, arg_masks, arg_shifts);
     }
 };
 
-#define DYNARMIC_DECODER_GET_MATCHER(MatcherT, fn, name, bitstring) Decoder::detail::detail<MatcherT<V>>::template GetMatcher<bitstring>(&V::fn)
+#define DYNARMIC_DECODER_GET_MATCHER(MatcherT, fn, name, bitstring) Decoder::detail::detail<MatcherT<V>>::GetMaskAndExpect(bitstring)
+
+#define DYNARMIC_DECODER_GET_MATCHER_FUNCTION(MatcherT, fn, name, bitstring) Decoder::detail::detail<MatcherT<V>>::template GetMatcherFunction<bitstring, V>(visitor, instruction, &V::fn)
 
 }  // namespace detail
 }  // namespace Dynarmic::Decoder
diff --git a/src/dynarmic/src/dynarmic/frontend/decoder/matcher.h b/src/dynarmic/src/dynarmic/frontend/decoder/matcher.h
index 194f646ed2..49095f67e3 100644
--- a/src/dynarmic/src/dynarmic/frontend/decoder/matcher.h
+++ b/src/dynarmic/src/dynarmic/frontend/decoder/matcher.h
@@ -15,50 +15,41 @@
 namespace Dynarmic::Decoder {
 
 /// Generic instruction handling construct.
-/// @tparam Visitor An arbitrary visitor type that will be passed through
-///                 to the function being handled. This type must be the
-///                 type of the first parameter in a handler function.
-/// @tparam OpcodeType Type representing an opcode. This must be the
-///                    type of the second parameter in a handler function.
-template<typename Visitor, typename OpcodeType>
+/// @tparam V An arbitrary visitor type that will be passed through
+/// to the function being handled. This type must be the
+/// type of the first parameter in a handler function.
+/// @tparam T Type representing an opcode. This must be the
+/// type of the second parameter in a handler function.
+template<typename V, typename T>
 class Matcher {
 public:
-    using opcode_type = OpcodeType;
-    using visitor_type = Visitor;
-    using handler_return_type = typename Visitor::instruction_return_type;
-    using handler_function = std::function<handler_return_type(Visitor&, opcode_type)>;
-    Matcher(opcode_type mask, opcode_type expected, handler_function func)
-            : mask{mask}, expected{expected}, fn{std::move(func)} {}
+    using opcode_type = T;
+    using visitor_type = V;
 
-    /// Gets the mask for this instruction.
-    inline opcode_type GetMask() const noexcept {
+    constexpr Matcher(T mask, T expected) noexcept
+        : mask{mask}
+        , expected{expected}
+    {}
+
+    /// @brief Gets the mask for this instruction.
+    constexpr inline T GetMask() const noexcept {
         return mask;
     }
 
-    /// Gets the expected value after masking for this instruction.
-    inline opcode_type GetExpected() const noexcept {
+    /// @brief Gets the expected value after masking for this instruction.
+    constexpr inline T GetExpected() const noexcept {
         return expected;
     }
 
-    /// Tests to see if the given instruction is the instruction this matcher represents.
+    /// @brief Tests to see if the given instruction is the instruction this matcher represents.
     /// @param instruction The instruction to test
     /// @returns true if the given instruction matches.
-    inline bool Matches(opcode_type instruction) const noexcept {
+    constexpr inline bool Matches(T instruction) const noexcept {
         return (instruction & mask) == expected;
     }
 
-    /// Calls the corresponding instruction handler on visitor for this type of instruction.
-    /// @param v The visitor to use
-    /// @param instruction The instruction to decode.
-    inline handler_return_type call(Visitor& v, opcode_type instruction) const noexcept {
-        ASSERT(Matches(instruction));
-        return fn(v, instruction);
-    }
-
-private:
-    opcode_type mask;
-    opcode_type expected;
-    handler_function fn;
+    T mask;
+    T expected;
 };
 
 }  // namespace Dynarmic::Decoder
diff --git a/src/dynarmic/src/dynarmic/frontend/imm.h b/src/dynarmic/src/dynarmic/frontend/imm.h
index 56fbcbda0e..f305b5bfc8 100644
--- a/src/dynarmic/src/dynarmic/frontend/imm.h
+++ b/src/dynarmic/src/dynarmic/frontend/imm.h
@@ -27,9 +27,8 @@ class Imm {
 public:
     static constexpr size_t bit_size = bit_size_;
 
-    explicit Imm(u32 value)
-            : value(value) {
-        ASSERT((mcl::bit::get_bits<0, bit_size - 1>(value) == value) && "More bits in value than expected");
+    explicit Imm(u32 value) : value(value) {
+        DEBUG_ASSERT((mcl::bit::get_bits<0, bit_size - 1>(value) == value) && "More bits in value than expected");
     }
 
     template<typename T = u32>
diff --git a/src/dynarmic/tests/decoder_tests.cpp b/src/dynarmic/tests/decoder_tests.cpp
index 2a6c0f6a4f..17ed2c0ce5 100644
--- a/src/dynarmic/tests/decoder_tests.cpp
+++ b/src/dynarmic/tests/decoder_tests.cpp
@@ -51,7 +51,7 @@ TEST_CASE("ASIMD Decoder: Ensure table order correctness", "[decode][a32][.]") {
 
             const bool iserr = is_decode_error(*iter, instruction);
             const auto alternative = std::find_if(table.cbegin(), iter, [instruction](const auto& m) {
-                return m.Matches(instruction);
+                return (instruction & mask) == expect;
             });
             const bool altiserr = is_decode_error(*alternative, instruction);
 
diff --git a/src/dynarmic/tests/print_info.cpp b/src/dynarmic/tests/print_info.cpp
index 6d6109d849..988a8d294e 100644
--- a/src/dynarmic/tests/print_info.cpp
+++ b/src/dynarmic/tests/print_info.cpp
@@ -43,12 +43,12 @@
 using namespace Dynarmic;
 
 std::string_view GetNameOfA32Instruction(u32 instruction) {
-    if (auto const vfp_decoder = A32::DecodeVFP<A32::TranslatorVisitor>(instruction))
-        return *A32::GetNameVFP<A32::TranslatorVisitor>(instruction);
-    else if (auto const asimd_decoder = A32::DecodeASIMD<A32::TranslatorVisitor>(instruction))
-        return *A32::GetNameASIMD<A32::TranslatorVisitor>(instruction);
-    else if (auto const decoder = A32::DecodeArm<A32::TranslatorVisitor>(instruction))
-        return *A32::GetNameARM<A32::TranslatorVisitor>(instruction);
+    if (auto const vfp_decoder = A32::GetNameVFP<A32::TranslatorVisitor>(instruction))
+        return *vfp_decoder;
+    else if (auto const asimd_decoder = A32::GetNameASIMD<A32::TranslatorVisitor>(instruction))
+        return *asimd_decoder;
+    else if (auto const decoder = A32::GetNameArm<A32::TranslatorVisitor>(instruction))
+        return *decoder;
     return "<null>";
 }