24 files changed, 227 insertions, 224 deletions

diff --git a/.ci/scripts/common/post-upload.sh b/.ci/scripts/common/post-upload.sh
index b80868635..e46ee0abb 100644
--- a/.ci/scripts/common/post-upload.sh
+++ b/.ci/scripts/common/post-upload.sh
@@ -4,9 +4,13 @@
 cp license.txt "$DIR_NAME"
 cp README.md "$DIR_NAME"
 
+tar -cJvf "${REV_NAME}-source.tar.xz" src externals CMakeLists.txt README.md license.txt
+cp "${REV_NAME}-source.tar.xz" "$DIR_NAME"
+
 tar $COMPRESSION_FLAGS "$ARCHIVE_NAME" "$DIR_NAME"
 
 mv "$DIR_NAME" $RELEASE_NAME
+mv "${REV_NAME}-source.tar.xz" $RELEASE_NAME
 
 7z a "$REV_NAME.7z" $RELEASE_NAME
 
diff --git a/.ci/scripts/windows/upload.ps1 b/.ci/scripts/windows/upload.ps1
index e34e6681f..b9b8b4af8 100644
--- a/.ci/scripts/windows/upload.ps1
+++ b/.ci/scripts/windows/upload.ps1
@@ -1,11 +1,12 @@
 param($BUILD_NAME)
 
-$GITDATE = $(git show -s --date=short --format='%ad') -replace "-",""
+$GITDATE = $(git show -s --date=short --format='%ad') -replace "-", ""
 $GITREV = $(git show -s --format='%h')
 
 if ("$BUILD_NAME" -eq "mainline") {
     $RELEASE_DIST = "yuzu-windows-msvc"
-} else {
+}
+else {
     $RELEASE_DIST = "yuzu-windows-msvc-$BUILD_NAME"
 }
 
@@ -14,6 +15,9 @@ $MSVC_BUILD_PDB = "yuzu-windows-msvc-$GITDATE-$GITREV-debugsymbols.zip" -replace
 $MSVC_SEVENZIP = "yuzu-windows-msvc-$GITDATE-$GITREV.7z" -replace " ", ""
 $MSVC_TAR = "yuzu-windows-msvc-$GITDATE-$GITREV.tar" -replace " ", ""
 $MSVC_TARXZ = "yuzu-windows-msvc-$GITDATE-$GITREV.tar.xz" -replace " ", ""
+$MSVC_SOURCE = "yuzu-windows-msvc-source-$GITDATE-$GITREV" -replace " ", ""
+$MSVC_SOURCE_TAR = "$MSVC_SOURCE.tar"
+$MSVC_SOURCE_TARXZ = "$MSVC_SOURCE_TAR.xz"
 
 $env:BUILD_ZIP = $MSVC_BUILD_ZIP
 $env:BUILD_SYMBOLS = $MSVC_BUILD_PDB
@@ -21,19 +25,33 @@ $env:BUILD_UPDATE = $MSVC_SEVENZIP
 
 $BUILD_DIR = ".\build\bin\Release"
 
+# Upload debugging symbols
 mkdir pdb
 Get-ChildItem "$BUILD_DIR\" -Recurse -Filter "*.pdb" | Copy-Item -destination .\pdb
 7z a -tzip $MSVC_BUILD_PDB .\pdb\*.pdb
 rm "$BUILD_DIR\*.pdb"
+
+# Create artifact directories
 mkdir $RELEASE_DIST
+mkdir $MSVC_SOURCE
 mkdir "artifacts"
 
+# Build a tar.xz for the source of the release
+Copy-Item .\license.txt -Destination $MSVC_SOURCE
+Copy-Item .\README.md -Destination $MSVC_SOURCE
+Copy-Item .\src -Recurse -Destination $MSVC_SOURCE
+Copy-Item .\externals -Recurse -Destination $MSVC_SOURCE
+Copy-Item .\dist -Recurse -Destination $MSVC_SOURCE
+Copy-Item .\CMakeModules -Recurse -Destination $MSVC_SOURCE
+7z a -r -ttar $MSVC_SOURCE_TAR $MSVC_SOURCE
+7z a -r -txz $MSVC_SOURCE_TARXZ $MSVC_SOURCE_TAR
+
+# Build the final release artifacts
+Copy-Item $MSVC_SOURCE_TARXZ -Destination $RELEASE_DIST
 Copy-Item "$BUILD_DIR\*" -Destination $RELEASE_DIST -Recurse
 rm "$RELEASE_DIST\*.exe"
 Get-ChildItem "$BUILD_DIR" -Recurse -Filter "yuzu*.exe" | Copy-Item -destination $RELEASE_DIST
 Get-ChildItem "$BUILD_DIR" -Recurse -Filter "QtWebEngineProcess*.exe" | Copy-Item -destination $RELEASE_DIST
-Copy-Item .\license.txt -Destination $RELEASE_DIST
-Copy-Item .\README.md -Destination $RELEASE_DIST
 7z a -tzip $MSVC_BUILD_ZIP $RELEASE_DIST\*
 7z a $MSVC_SEVENZIP $RELEASE_DIST
 
diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md
index ccb8fc5d7..5d4b6f9da 100644
--- a/CONTRIBUTING.md
+++ b/CONTRIBUTING.md
@@ -1 +1 @@
-**The Contributor's Guide has moved to [the Yuzu wiki](https://github.com/yuzu-emu/yuzu/wiki/Contributing).**
+**The Contributor's Guide has moved to [the yuzu wiki](https://github.com/yuzu-emu/yuzu/wiki/Contributing).**
diff --git a/externals/microprofile/microprofile.h b/externals/microprofile/microprofile.h
index 384863ccc..cdb312b87 100644
--- a/externals/microprofile/microprofile.h
+++ b/externals/microprofile/microprofile.h
@@ -814,7 +814,7 @@ struct MicroProfile
 
 inline int MicroProfileLogType(MicroProfileLogEntry Index)
 {
-    return ((MP_LOG_BEGIN_MASK & Index)>>62) & 0x3;
+    return (int)(((MP_LOG_BEGIN_MASK & Index)>>62) & 0x3ULL);
 }
 
 inline uint64_t MicroProfileLogTimerIndex(MicroProfileLogEntry Index)
@@ -861,12 +861,12 @@ T MicroProfileMax(T a, T b)
 
 inline int64_t MicroProfileMsToTick(float fMs, int64_t nTicksPerSecond)
 {
-    return (int64_t)(fMs*0.001f*nTicksPerSecond);
+    return (int64_t)(fMs*0.001f*(float)nTicksPerSecond);
 }
 
 inline float MicroProfileTickToMsMultiplier(int64_t nTicksPerSecond)
 {
-    return 1000.f / nTicksPerSecond;
+    return 1000.f / (float)nTicksPerSecond;
 }
 
 inline uint16_t MicroProfileGetGroupIndex(MicroProfileToken t)
diff --git a/src/common/common_funcs.h b/src/common/common_funcs.h
index c029dc7b3..6dc3e108f 100644
--- a/src/common/common_funcs.h
+++ b/src/common/common_funcs.h
@@ -56,7 +56,7 @@ std::string GetLastErrorMsg();
 namespace Common {
 
 constexpr u32 MakeMagic(char a, char b, char c, char d) {
-    return a | b << 8 | c << 16 | d << 24;
+    return u32(a) | u32(b) << 8 | u32(c) << 16 | u32(d) << 24;
 }
 
 } // namespace Common
diff --git a/src/common/hash.h b/src/common/hash.h
index ebd4125e2..b2538f3ea 100644
--- a/src/common/hash.h
+++ b/src/common/hash.h
@@ -35,41 +35,6 @@ static inline u64 ComputeStructHash64(const T& data) {
     return ComputeHash64(&data, sizeof(data));
 }
 
-/// A helper template that ensures the padding in a struct is initialized by memsetting to 0.
-template <typename T>
-struct HashableStruct {
-    // In addition to being trivially copyable, T must also have a trivial default constructor,
-    // because any member initialization would be overridden by memset
-    static_assert(std::is_trivial_v<T>, "Type passed to HashableStruct must be trivial");
-    /*
-     * We use a union because "implicitly-defined copy/move constructor for a union X copies the
-     * object representation of X." and "implicitly-defined copy assignment operator for a union X
-     * copies the object representation (3.9) of X." = Bytewise copy instead of memberwise copy.
-     * This is important because the padding bytes are included in the hash and comparison between
-     * objects.
-     */
-    union {
-        T state;
-    };
-
-    HashableStruct() {
-        // Memset structure to zero padding bits, so that they will be deterministic when hashing
-        std::memset(&state, 0, sizeof(T));
-    }
-
-    bool operator==(const HashableStruct<T>& o) const {
-        return std::memcmp(&state, &o.state, sizeof(T)) == 0;
-    };
-
-    bool operator!=(const HashableStruct<T>& o) const {
-        return !(*this == o);
-    };
-
-    std::size_t Hash() const {
-        return Common::ComputeStructHash64(state);
-    }
-};
-
 struct PairHash {
     template <class T1, class T2>
     std::size_t operator()(const std::pair<T1, T2>& pair) const noexcept {
diff --git a/src/core/file_sys/xts_archive.cpp b/src/core/file_sys/xts_archive.cpp
index 4bc5cb2ee..f5f8b91c9 100644
--- a/src/core/file_sys/xts_archive.cpp
+++ b/src/core/file_sys/xts_archive.cpp
@@ -93,8 +93,7 @@ Loader::ResultStatus NAX::Parse(std::string_view path) {
     std::size_t i = 0;
     for (; i < sd_keys.size(); ++i) {
         std::array<Core::Crypto::Key128, 2> nax_keys{};
-        if (!CalculateHMAC256(nax_keys.data(), sd_keys[i].data(), 0x10, std::string(path).c_str(),
-                              path.size())) {
+        if (!CalculateHMAC256(nax_keys.data(), sd_keys[i].data(), 0x10, path.data(), path.size())) {
             return Loader::ResultStatus::ErrorNAXKeyHMACFailed;
         }
 
diff --git a/src/video_core/CMakeLists.txt b/src/video_core/CMakeLists.txt
index c911c6ec4..45d8eaf23 100644
--- a/src/video_core/CMakeLists.txt
+++ b/src/video_core/CMakeLists.txt
@@ -180,3 +180,9 @@ target_link_libraries(video_core PRIVATE glad)
 if (ENABLE_VULKAN)
     target_link_libraries(video_core PRIVATE sirit)
 endif()
+
+if (MSVC)
+    target_compile_options(video_core PRIVATE /we4267)
+else()
+    target_compile_options(video_core PRIVATE -Werror=conversion -Wno-error=sign-conversion)
+endif()
diff --git a/src/video_core/engines/maxwell_3d.cpp b/src/video_core/engines/maxwell_3d.cpp
index 2bed6cb38..42ce49a4d 100644
--- a/src/video_core/engines/maxwell_3d.cpp
+++ b/src/video_core/engines/maxwell_3d.cpp
@@ -261,7 +261,8 @@ void Maxwell3D::CallMacroMethod(u32 method, std::size_t num_parameters, const u3
     executing_macro = 0;
 
     // Lookup the macro offset
-    const u32 entry = ((method - MacroRegistersStart) >> 1) % macro_positions.size();
+    const u32 entry =
+        ((method - MacroRegistersStart) >> 1) % static_cast<u32>(macro_positions.size());
 
     // Execute the current macro.
     macro_interpreter.Execute(macro_positions[entry], num_parameters, parameters);
diff --git a/src/video_core/engines/shader_bytecode.h b/src/video_core/engines/shader_bytecode.h
index 8f6bc76eb..9fafed4a2 100644
--- a/src/video_core/engines/shader_bytecode.h
+++ b/src/video_core/engines/shader_bytecode.h
@@ -616,6 +616,14 @@ union Instruction {
     } shfl;
 
     union {
+        BitField<44, 1, u64> ftz;
+        BitField<39, 2, u64> tab5cb8_2;
+        BitField<38, 1, u64> ndv;
+        BitField<47, 1, u64> cc;
+        BitField<28, 8, u64> swizzle;
+    } fswzadd;
+
+    union {
         BitField<8, 8, Register> gpr;
         BitField<20, 24, s64> offset;
     } gmem;
@@ -1478,7 +1486,8 @@ union Instruction {
             u32 value = static_cast<u32>(target);
             // The branch offset is relative to the next instruction and is stored in bytes, so
             // divide it by the size of an instruction and add 1 to it.
-            return static_cast<s32>((value ^ mask) - mask) / sizeof(Instruction) + 1;
+            return static_cast<s32>((value ^ mask) - mask) / static_cast<s32>(sizeof(Instruction)) +
+                   1;
         }
     } bra;
 
@@ -1492,7 +1501,8 @@ union Instruction {
             u32 value = static_cast<u32>(target);
             // The branch offset is relative to the next instruction and is stored in bytes, so
             // divide it by the size of an instruction and add 1 to it.
-            return static_cast<s32>((value ^ mask) - mask) / sizeof(Instruction) + 1;
+            return static_cast<s32>((value ^ mask) - mask) / static_cast<s32>(sizeof(Instruction)) +
+                   1;
         }
     } brx;
 
@@ -1590,6 +1600,7 @@ public:
         DEPBAR,
         VOTE,
         SHFL,
+        FSWZADD,
         BFE_C,
         BFE_R,
         BFE_IMM,
@@ -1851,11 +1862,11 @@ private:
                 const std::size_t bit_position = opcode_bitsize - i - 1;
                 switch (bitstring[i]) {
                 case '0':
-                    mask |= 1 << bit_position;
+                    mask |= static_cast<u16>(1U << bit_position);
                     break;
                 case '1':
-                    expect |= 1 << bit_position;
-                    mask |= 1 << bit_position;
+                    expect |= static_cast<u16>(1U << bit_position);
+                    mask |= static_cast<u16>(1U << bit_position);
                     break;
                 default:
                     // Ignore
@@ -1888,6 +1899,7 @@ private:
             INST("1111000011110---", Id::DEPBAR, Type::Synch, "DEPBAR"),
             INST("0101000011011---", Id::VOTE, Type::Warp, "VOTE"),
             INST("1110111100010---", Id::SHFL, Type::Warp, "SHFL"),
+            INST("0101000011111---", Id::FSWZADD, Type::Warp, "FSWZADD"),
             INST("1110111111011---", Id::LD_A, Type::Memory, "LD_A"),
             INST("1110111101001---", Id::LD_S, Type::Memory, "LD_S"),
             INST("1110111101000---", Id::LD_L, Type::Memory, "LD_L"),
diff --git a/src/video_core/renderer_opengl/gl_device.cpp b/src/video_core/renderer_opengl/gl_device.cpp
index c65b24c69..b30d5be74 100644
--- a/src/video_core/renderer_opengl/gl_device.cpp
+++ b/src/video_core/renderer_opengl/gl_device.cpp
@@ -62,6 +62,7 @@ Device::Device() {
     max_varyings = GetInteger<u32>(GL_MAX_VARYING_VECTORS);
     has_warp_intrinsics = GLAD_GL_NV_gpu_shader5 && GLAD_GL_NV_shader_thread_group &&
                           GLAD_GL_NV_shader_thread_shuffle;
+    has_shader_ballot = GLAD_GL_ARB_shader_ballot;
     has_vertex_viewport_layer = GLAD_GL_ARB_shader_viewport_layer_array;
     has_image_load_formatted = HasExtension(extensions, "GL_EXT_shader_image_load_formatted");
     has_variable_aoffi = TestVariableAoffi();
@@ -79,6 +80,7 @@ Device::Device(std::nullptr_t) {
     max_vertex_attributes = 16;
     max_varyings = 15;
     has_warp_intrinsics = true;
+    has_shader_ballot = true;
     has_vertex_viewport_layer = true;
     has_image_load_formatted = true;
     has_variable_aoffi = true;
diff --git a/src/video_core/renderer_opengl/gl_device.h b/src/video_core/renderer_opengl/gl_device.h
index bf35bd0b6..6c86fe207 100644
--- a/src/video_core/renderer_opengl/gl_device.h
+++ b/src/video_core/renderer_opengl/gl_device.h
@@ -34,6 +34,10 @@ public:
         return has_warp_intrinsics;
     }
 
+    bool HasShaderBallot() const {
+        return has_shader_ballot;
+    }
+
     bool HasVertexViewportLayer() const {
         return has_vertex_viewport_layer;
     }
@@ -68,6 +72,7 @@ private:
     u32 max_vertex_attributes{};
     u32 max_varyings{};
     bool has_warp_intrinsics{};
+    bool has_shader_ballot{};
     bool has_vertex_viewport_layer{};
     bool has_image_load_formatted{};
     bool has_variable_aoffi{};
diff --git a/src/video_core/renderer_opengl/gl_rasterizer.cpp b/src/video_core/renderer_opengl/gl_rasterizer.cpp
index e560d70d5..e43ba9d6b 100644
--- a/src/video_core/renderer_opengl/gl_rasterizer.cpp
+++ b/src/video_core/renderer_opengl/gl_rasterizer.cpp
@@ -375,7 +375,7 @@ void RasterizerOpenGL::ConfigureFramebuffers() {
         fbkey.color_attachments[index] = GL_COLOR_ATTACHMENT0 + regs.rt_control.GetMap(index);
         fbkey.colors[index] = std::move(color_surface);
     }
-    fbkey.colors_count = regs.rt_control.count;
+    fbkey.colors_count = static_cast<u16>(regs.rt_control.count);
 
     if (depth_surface) {
         // Assume that a surface will be written to if it is used as a framebuffer, even if
diff --git a/src/video_core/renderer_opengl/gl_shader_cache.cpp b/src/video_core/renderer_opengl/gl_shader_cache.cpp
index 35e5214a5..04a239a39 100644
--- a/src/video_core/renderer_opengl/gl_shader_cache.cpp
+++ b/src/video_core/renderer_opengl/gl_shader_cache.cpp
@@ -275,16 +275,25 @@ CachedProgram BuildShader(const Device& device, u64 unique_identifier, ProgramTy
     std::string source = fmt::format(R"(// {}
 #version 430 core
 #extension GL_ARB_separate_shader_objects : enable
-#extension GL_ARB_shader_viewport_layer_array : enable
-#extension GL_EXT_shader_image_load_formatted : enable
-#extension GL_NV_gpu_shader5 : enable
-#extension GL_NV_shader_thread_group : enable
-#extension GL_NV_shader_thread_shuffle : enable
 )",
                                      GetShaderId(unique_identifier, program_type));
     if (is_compute) {
         source += "#extension GL_ARB_compute_variable_group_size : require\n";
     }
+    if (device.HasShaderBallot()) {
+        source += "#extension GL_ARB_shader_ballot : require\n";
+    }
+    if (device.HasVertexViewportLayer()) {
+        source += "#extension GL_ARB_shader_viewport_layer_array : require\n";
+    }
+    if (device.HasImageLoadFormatted()) {
+        source += "#extension GL_EXT_shader_image_load_formatted : require\n";
+    }
+    if (device.HasWarpIntrinsics()) {
+        source += "#extension GL_NV_gpu_shader5 : require\n"
+                  "#extension GL_NV_shader_thread_group : require\n"
+                  "#extension GL_NV_shader_thread_shuffle : require\n";
+    }
     source += '\n';
 
     if (!is_compute) {
diff --git a/src/video_core/renderer_opengl/gl_shader_decompiler.cpp b/src/video_core/renderer_opengl/gl_shader_decompiler.cpp
index 0ce59a852..e56ed51de 100644
--- a/src/video_core/renderer_opengl/gl_shader_decompiler.cpp
+++ b/src/video_core/renderer_opengl/gl_shader_decompiler.cpp
@@ -1379,6 +1379,26 @@ private:
         return GenerateUnary(operation, "float", Type::Float, type);
     }
 
+    Expression FSwizzleAdd(Operation operation) {
+        const std::string op_a = VisitOperand(operation, 0).AsFloat();
+        const std::string op_b = VisitOperand(operation, 1).AsFloat();
+
+        if (!device.HasShaderBallot()) {
+            LOG_ERROR(Render_OpenGL, "Shader ballot is unavailable but required by the shader");
+            return {fmt::format("{} + {}", op_a, op_b), Type::Float};
+        }
+
+        const std::string instr_mask = VisitOperand(operation, 2).AsUint();
+        const std::string mask = code.GenerateTemporary();
+        code.AddLine("uint {} = ({} >> ((gl_SubGroupInvocationARB & 3) << 1)) & 3;", mask,
+                     instr_mask);
+
+        const std::string modifier_a = fmt::format("fswzadd_modifiers_a[{}]", mask);
+        const std::string modifier_b = fmt::format("fswzadd_modifiers_b[{}]", mask);
+        return {fmt::format("(({} * {}) + ({} * {}))", op_a, modifier_a, op_b, modifier_b),
+                Type::Float};
+    }
+
     Expression ICastFloat(Operation operation) {
         return GenerateUnary(operation, "int", Type::Int, Type::Float);
     }
@@ -1942,34 +1962,24 @@ private:
         return Vote(operation, "allThreadsEqualNV");
     }
 
-    template <const std::string_view& func>
-    Expression Shuffle(Operation operation) {
-        const std::string value = VisitOperand(operation, 0).AsFloat();
-        if (!device.HasWarpIntrinsics()) {
-            LOG_ERROR(Render_OpenGL, "Nvidia shuffle intrinsics are required by this shader");
-            // On a "single-thread" device we are either on the same thread or out of bounds. Both
-            // cases return the passed value.
-            return {value, Type::Float};
+    Expression ThreadId(Operation operation) {
+        if (!device.HasShaderBallot()) {
+            LOG_ERROR(Render_OpenGL, "Shader ballot is unavailable but required by the shader");
+            return {"0U", Type::Uint};
         }
-
-        const std::string index = VisitOperand(operation, 1).AsUint();
-        const std::string width = VisitOperand(operation, 2).AsUint();
-        return {fmt::format("{}({}, {}, {})", func, value, index, width), Type::Float};
+        return {"gl_SubGroupInvocationARB", Type::Uint};
     }
 
-    template <const std::string_view& func>
-    Expression InRangeShuffle(Operation operation) {
-        const std::string index = VisitOperand(operation, 0).AsUint();
-        const std::string width = VisitOperand(operation, 1).AsUint();
-        if (!device.HasWarpIntrinsics()) {
-            // On a "single-thread" device we are only in bounds when the requested index is 0.
-            return {fmt::format("({} == 0U)", index), Type::Bool};
+    Expression ShuffleIndexed(Operation operation) {
+        std::string value = VisitOperand(operation, 0).AsFloat();
+
+        if (!device.HasShaderBallot()) {
+            LOG_ERROR(Render_OpenGL, "Shader ballot is unavailable but required by the shader");
+            return {std::move(value), Type::Float};
         }
 
-        const std::string in_range = code.GenerateTemporary();
-        code.AddLine("bool {};", in_range);
-        code.AddLine("{}(0U, {}, {}, {});", func, index, width, in_range);
-        return {in_range, Type::Bool};
+        const std::string index = VisitOperand(operation, 1).AsUint();
+        return {fmt::format("readInvocationARB({}, {})", value, index), Type::Float};
     }
 
     struct Func final {
@@ -1981,11 +1991,6 @@ private:
         static constexpr std::string_view Or = "Or";
         static constexpr std::string_view Xor = "Xor";
         static constexpr std::string_view Exchange = "Exchange";
-
-        static constexpr std::string_view ShuffleIndexed = "shuffleNV";
-        static constexpr std::string_view ShuffleUp = "shuffleUpNV";
-        static constexpr std::string_view ShuffleDown = "shuffleDownNV";
-        static constexpr std::string_view ShuffleButterfly = "shuffleXorNV";
     };
 
     static constexpr std::array operation_decompilers = {
@@ -2016,6 +2021,7 @@ private:
         &GLSLDecompiler::FTrunc,
         &GLSLDecompiler::FCastInteger<Type::Int>,
         &GLSLDecompiler::FCastInteger<Type::Uint>,
+        &GLSLDecompiler::FSwizzleAdd,
 
         &GLSLDecompiler::Add<Type::Int>,
         &GLSLDecompiler::Mul<Type::Int>,
@@ -2151,15 +2157,8 @@ private:
         &GLSLDecompiler::VoteAny,
         &GLSLDecompiler::VoteEqual,
 
-        &GLSLDecompiler::Shuffle<Func::ShuffleIndexed>,
-        &GLSLDecompiler::Shuffle<Func::ShuffleUp>,
-        &GLSLDecompiler::Shuffle<Func::ShuffleDown>,
-        &GLSLDecompiler::Shuffle<Func::ShuffleButterfly>,
-
-        &GLSLDecompiler::InRangeShuffle<Func::ShuffleIndexed>,
-        &GLSLDecompiler::InRangeShuffle<Func::ShuffleUp>,
-        &GLSLDecompiler::InRangeShuffle<Func::ShuffleDown>,
-        &GLSLDecompiler::InRangeShuffle<Func::ShuffleButterfly>,
+        &GLSLDecompiler::ThreadId,
+        &GLSLDecompiler::ShuffleIndexed,
     };
     static_assert(operation_decompilers.size() == static_cast<std::size_t>(OperationCode::Amount));
 
@@ -2492,6 +2491,9 @@ bvec2 HalfFloatNanComparison(bvec2 comparison, vec2 pair1, vec2 pair2) {
     bvec2 is_nan2 = isnan(pair2);
     return bvec2(comparison.x || is_nan1.x || is_nan2.x, comparison.y || is_nan1.y || is_nan2.y);
 }
+
+const float fswzadd_modifiers_a[] = float[4](-1.0f,  1.0f, -1.0f,  0.0f );
+const float fswzadd_modifiers_b[] = float[4](-1.0f, -1.0f,  1.0f, -1.0f );
 )";
 }
 
diff --git a/src/video_core/renderer_opengl/renderer_opengl.cpp b/src/video_core/renderer_opengl/renderer_opengl.cpp
index 4bbd17b12..7646cbb0e 100644
--- a/src/video_core/renderer_opengl/renderer_opengl.cpp
+++ b/src/video_core/renderer_opengl/renderer_opengl.cpp
@@ -323,10 +323,12 @@ void RendererOpenGL::DrawScreenTriangles(const ScreenInfo& screen_info, float x,
     // (e.g. handheld mode) on a 1920x1080 framebuffer.
     f32 scale_u = 1.f, scale_v = 1.f;
     if (framebuffer_crop_rect.GetWidth() > 0) {
-        scale_u = static_cast<f32>(framebuffer_crop_rect.GetWidth()) / screen_info.texture.width;
+        scale_u = static_cast<f32>(framebuffer_crop_rect.GetWidth()) /
+                  static_cast<f32>(screen_info.texture.width);
     }
     if (framebuffer_crop_rect.GetHeight() > 0) {
-        scale_v = static_cast<f32>(framebuffer_crop_rect.GetHeight()) / screen_info.texture.height;
+        scale_v = static_cast<f32>(framebuffer_crop_rect.GetHeight()) /
+                  static_cast<f32>(screen_info.texture.height);
     }
 
     std::array<ScreenRectVertex, 4> vertices = {{
diff --git a/src/video_core/renderer_vulkan/vk_shader_decompiler.cpp b/src/video_core/renderer_vulkan/vk_shader_decompiler.cpp
index 42cf068b6..2850d5b59 100644
--- a/src/video_core/renderer_vulkan/vk_shader_decompiler.cpp
+++ b/src/video_core/renderer_vulkan/vk_shader_decompiler.cpp
@@ -783,6 +783,11 @@ private:
         return {};
     }
 
+    Id FSwizzleAdd(Operation operation) {
+        UNIMPLEMENTED();
+        return {};
+    }
+
     Id HNegate(Operation operation) {
         UNIMPLEMENTED();
         return {};
@@ -1195,42 +1200,12 @@ private:
         return {};
     }
 
-    Id ShuffleIndexed(Operation) {
-        UNIMPLEMENTED();
-        return {};
-    }
-
-    Id ShuffleUp(Operation) {
-        UNIMPLEMENTED();
-        return {};
-    }
-
-    Id ShuffleDown(Operation) {
-        UNIMPLEMENTED();
-        return {};
-    }
-
-    Id ShuffleButterfly(Operation) {
-        UNIMPLEMENTED();
-        return {};
-    }
-
-    Id InRangeShuffleIndexed(Operation) {
+    Id ThreadId(Operation) {
         UNIMPLEMENTED();
         return {};
     }
 
-    Id InRangeShuffleUp(Operation) {
-        UNIMPLEMENTED();
-        return {};
-    }
-
-    Id InRangeShuffleDown(Operation) {
-        UNIMPLEMENTED();
-        return {};
-    }
-
-    Id InRangeShuffleButterfly(Operation) {
+    Id ShuffleIndexed(Operation) {
         UNIMPLEMENTED();
         return {};
     }
@@ -1393,6 +1368,7 @@ private:
         &SPIRVDecompiler::Unary<&Module::OpTrunc, Type::Float>,
         &SPIRVDecompiler::Unary<&Module::OpConvertSToF, Type::Float, Type::Int>,
         &SPIRVDecompiler::Unary<&Module::OpConvertUToF, Type::Float, Type::Uint>,
+        &SPIRVDecompiler::FSwizzleAdd,
 
         &SPIRVDecompiler::Binary<&Module::OpIAdd, Type::Int>,
         &SPIRVDecompiler::Binary<&Module::OpIMul, Type::Int>,
@@ -1528,15 +1504,8 @@ private:
         &SPIRVDecompiler::VoteAny,
         &SPIRVDecompiler::VoteEqual,
 
+        &SPIRVDecompiler::ThreadId,
         &SPIRVDecompiler::ShuffleIndexed,
-        &SPIRVDecompiler::ShuffleUp,
-        &SPIRVDecompiler::ShuffleDown,
-        &SPIRVDecompiler::ShuffleButterfly,
-
-        &SPIRVDecompiler::InRangeShuffleIndexed,
-        &SPIRVDecompiler::InRangeShuffleUp,
-        &SPIRVDecompiler::InRangeShuffleDown,
-        &SPIRVDecompiler::InRangeShuffleButterfly,
     };
     static_assert(operation_decompilers.size() == static_cast<std::size_t>(OperationCode::Amount));
 
diff --git a/src/video_core/shader/decode.cpp b/src/video_core/shader/decode.cpp
index 21fb9cb83..22c3e5120 100644
--- a/src/video_core/shader/decode.cpp
+++ b/src/video_core/shader/decode.cpp
@@ -154,10 +154,10 @@ void ShaderIR::Decode() {
         LOG_CRITICAL(HW_GPU, "Unknown decompilation mode!");
         [[fallthrough]];
     case CompileDepth::BruteForce: {
+        const auto shader_end = static_cast<u32>(program_code.size());
         coverage_begin = main_offset;
-        const std::size_t shader_end = program_code.size();
         coverage_end = shader_end;
-        for (u32 label = main_offset; label < shader_end; label++) {
+        for (u32 label = main_offset; label < shader_end; ++label) {
             basic_blocks.insert({label, DecodeRange(label, label + 1)});
         }
         break;
diff --git a/src/video_core/shader/decode/warp.cpp b/src/video_core/shader/decode/warp.cpp
index fa8a250cc..d98d0e1dd 100644
--- a/src/video_core/shader/decode/warp.cpp
+++ b/src/video_core/shader/decode/warp.cpp
@@ -17,6 +17,7 @@ using Tegra::Shader::ShuffleOperation;
 using Tegra::Shader::VoteOperation;
 
 namespace {
+
 OperationCode GetOperationCode(VoteOperation vote_op) {
     switch (vote_op) {
     case VoteOperation::All:
@@ -30,6 +31,7 @@ OperationCode GetOperationCode(VoteOperation vote_op) {
         return OperationCode::VoteAll;
     }
 }
+
 } // Anonymous namespace
 
 u32 ShaderIR::DecodeWarp(NodeBlock& bb, u32 pc) {
@@ -46,50 +48,59 @@ u32 ShaderIR::DecodeWarp(NodeBlock& bb, u32 pc) {
         break;
     }
     case OpCode::Id::SHFL: {
-        Node width = [this, instr] {
-            Node mask = instr.shfl.is_mask_imm ? Immediate(static_cast<u32>(instr.shfl.mask_imm))
-                                               : GetRegister(instr.gpr39);
-
-            // Convert the obscure SHFL mask back into GL_NV_shader_thread_shuffle's width. This has
-            // been done reversing Nvidia's math. It won't work on all cases due to SHFL having
-            // different parameters that don't properly map to GLSL's interface, but it should work
-            // for cases emitted by Nvidia's compiler.
-            if (instr.shfl.operation == ShuffleOperation::Up) {
-                return Operation(
-                    OperationCode::ILogicalShiftRight,
-                    Operation(OperationCode::IAdd, std::move(mask), Immediate(-0x2000)),
-                    Immediate(8));
-            } else {
-                return Operation(OperationCode::ILogicalShiftRight,
-                                 Operation(OperationCode::IAdd, Immediate(0x201F),
-                                           Operation(OperationCode::INegate, std::move(mask))),
-                                 Immediate(8));
-            }
-        }();
+        Node mask = instr.shfl.is_mask_imm ? Immediate(static_cast<u32>(instr.shfl.mask_imm))
+                                           : GetRegister(instr.gpr39);
+        Node index = instr.shfl.is_index_imm ? Immediate(static_cast<u32>(instr.shfl.index_imm))
+                                             : GetRegister(instr.gpr20);
+
+        Node thread_id = Operation(OperationCode::ThreadId);
+        Node clamp = Operation(OperationCode::IBitwiseAnd, mask, Immediate(0x1FU));
+        Node seg_mask = BitfieldExtract(mask, 8, 16);
 
-        const auto [operation, in_range] = [instr]() -> std::pair<OperationCode, OperationCode> {
+        Node neg_seg_mask = Operation(OperationCode::IBitwiseNot, seg_mask);
+        Node min_thread_id = Operation(OperationCode::IBitwiseAnd, thread_id, seg_mask);
+        Node max_thread_id = Operation(OperationCode::IBitwiseOr, min_thread_id,
+                                       Operation(OperationCode::IBitwiseAnd, clamp, neg_seg_mask));
+
+        Node src_thread_id = [instr, index, neg_seg_mask, min_thread_id, thread_id] {
             switch (instr.shfl.operation) {
             case ShuffleOperation::Idx:
-                return {OperationCode::ShuffleIndexed, OperationCode::InRangeShuffleIndexed};
-            case ShuffleOperation::Up:
-                return {OperationCode::ShuffleUp, OperationCode::InRangeShuffleUp};
+                return Operation(OperationCode::IBitwiseOr,
+                                 Operation(OperationCode::IBitwiseAnd, index, neg_seg_mask),
+                                 min_thread_id);
             case ShuffleOperation::Down:
-                return {OperationCode::ShuffleDown, OperationCode::InRangeShuffleDown};
+                return Operation(OperationCode::IAdd, thread_id, index);
+            case ShuffleOperation::Up:
+                return Operation(OperationCode::IAdd, thread_id,
+                                 Operation(OperationCode::INegate, index));
             case ShuffleOperation::Bfly:
-                return {OperationCode::ShuffleButterfly, OperationCode::InRangeShuffleButterfly};
+                return Operation(OperationCode::IBitwiseXor, thread_id, index);
             }
-            UNREACHABLE_MSG("Invalid SHFL operation: {}",
-                            static_cast<u64>(instr.shfl.operation.Value()));
-            return {};
+            UNREACHABLE();
+            return Immediate(0U);
         }();
 
-        // Setting the predicate before the register is intentional to avoid overwriting.
-        Node index = instr.shfl.is_index_imm ? Immediate(static_cast<u32>(instr.shfl.index_imm))
-                                             : GetRegister(instr.gpr20);
-        SetPredicate(bb, instr.shfl.pred48, Operation(in_range, index, width));
+        Node in_bounds = [instr, src_thread_id, min_thread_id, max_thread_id] {
+            if (instr.shfl.operation == ShuffleOperation::Up) {
+                return Operation(OperationCode::LogicalIGreaterEqual, src_thread_id, min_thread_id);
+            } else {
+                return Operation(OperationCode::LogicalILessEqual, src_thread_id, max_thread_id);
+            }
+        }();
+
+        SetPredicate(bb, instr.shfl.pred48, in_bounds);
         SetRegister(
             bb, instr.gpr0,
-            Operation(operation, GetRegister(instr.gpr8), std::move(index), std::move(width)));
+            Operation(OperationCode::ShuffleIndexed, GetRegister(instr.gpr8), src_thread_id));
+        break;
+    }
+    case OpCode::Id::FSWZADD: {
+        UNIMPLEMENTED_IF(instr.fswzadd.ndv);
+
+        Node op_a = GetRegister(instr.gpr8);
+        Node op_b = GetRegister(instr.gpr20);
+        Node mask = Immediate(static_cast<u32>(instr.fswzadd.swizzle));
+        SetRegister(bb, instr.gpr0, Operation(OperationCode::FSwizzleAdd, op_a, op_b, mask));
         break;
     }
     default:
diff --git a/src/video_core/shader/node.h b/src/video_core/shader/node.h
index 4300d9ff4..54217e6a4 100644
--- a/src/video_core/shader/node.h
+++ b/src/video_core/shader/node.h
@@ -47,6 +47,7 @@ enum class OperationCode {
     FTrunc,        /// (MetaArithmetic, float a) -> float
     FCastInteger,  /// (MetaArithmetic, int a) -> float
     FCastUInteger, /// (MetaArithmetic, uint a) -> float
+    FSwizzleAdd,   /// (float a, float b, uint mask) -> float
 
     IAdd,                  /// (MetaArithmetic, int a, int b) -> int
     IMul,                  /// (MetaArithmetic, int a, int b) -> int
@@ -181,15 +182,8 @@ enum class OperationCode {
     VoteAny,      /// (bool) -> bool
     VoteEqual,    /// (bool) -> bool
 
-    ShuffleIndexed,   /// (uint value, uint index, uint width) -> uint
-    ShuffleUp,        /// (uint value, uint index, uint width) -> uint
-    ShuffleDown,      /// (uint value, uint index, uint width) -> uint
-    ShuffleButterfly, /// (uint value, uint index, uint width) -> uint
-
-    InRangeShuffleIndexed,   /// (uint index, uint width) -> bool
-    InRangeShuffleUp,        /// (uint index, uint width) -> bool
-    InRangeShuffleDown,      /// (uint index, uint width) -> bool
-    InRangeShuffleButterfly, /// (uint index, uint width) -> bool
+    ThreadId,       /// () -> uint
+    ShuffleIndexed, /// (uint value, uint index) -> uint
 
     Amount,
 };
diff --git a/src/video_core/shader/shader_ir.h b/src/video_core/shader/shader_ir.h
index 26c8fde22..76a849818 100644
--- a/src/video_core/shader/shader_ir.h
+++ b/src/video_core/shader/shader_ir.h
@@ -49,7 +49,7 @@ public:
     }
 
     u32 GetSize() const {
-        return max_offset + sizeof(float);
+        return max_offset + static_cast<u32>(sizeof(float));
     }
 
     u32 GetMaxOffset() const {
@@ -165,8 +165,8 @@ public:
         return program_manager.GetVariables();
     }
 
-    u32 ConvertAddressToNvidiaSpace(const u32 address) const {
-        return (address - main_offset) * sizeof(Tegra::Shader::Instruction);
+    u32 ConvertAddressToNvidiaSpace(u32 address) const {
+        return (address - main_offset) * static_cast<u32>(sizeof(Tegra::Shader::Instruction));
     }
 
     /// Returns a condition code evaluated from internal flags
diff --git a/src/video_core/textures/astc.cpp b/src/video_core/textures/astc.cpp
index 58b608a36..33bd31865 100644
--- a/src/video_core/textures/astc.cpp
+++ b/src/video_core/textures/astc.cpp
@@ -92,11 +92,11 @@ private:
         const unsigned int mask = 1 << m_NextBit++;
 
         // clear the bit
-        *m_CurByte &= ~mask;
+        *m_CurByte &= static_cast<unsigned char>(~mask);
 
         // Write the bit, if necessary
         if (b)
-            *m_CurByte |= mask;
+            *m_CurByte |= static_cast<unsigned char>(mask);
 
         // Next byte?
         if (m_NextBit >= 8) {
@@ -137,7 +137,7 @@ public:
         }
 
         uint64_t mask = (1 << (end - start + 1)) - 1;
-        return (m_Bits >> start) & mask;
+        return (m_Bits >> start) & static_cast<IntType>(mask);
     }
 
 private:
@@ -656,7 +656,7 @@ static IntType Replicate(const IntType& val, uint32_t numBits, uint32_t toBit) {
         return 0;
     if (toBit == 0)
         return 0;
-    IntType v = val & ((1 << numBits) - 1);
+    IntType v = val & static_cast<IntType>((1 << numBits) - 1);
     IntType res = v;
     uint32_t reslen = numBits;
     while (reslen < toBit) {
@@ -666,8 +666,8 @@ static IntType Replicate(const IntType& val, uint32_t numBits, uint32_t toBit) {
             comp = numBits - newshift;
             numBits = newshift;
         }
-        res <<= numBits;
-        res |= v >> comp;
+        res = static_cast<IntType>(res << numBits);
+        res = static_cast<IntType>(res | (v >> comp));
         reslen += numBits;
     }
     return res;
@@ -714,7 +714,7 @@ public:
             // Do nothing
             return val;
         } else if (oldDepth == 0 && newDepth != 0) {
-            return (1 << newDepth) - 1;
+            return static_cast<ChannelType>((1 << newDepth) - 1);
         } else if (newDepth > oldDepth) {
             return Replicate(val, oldDepth, newDepth);
         } else {
@@ -722,10 +722,11 @@ public:
             if (newDepth == 0) {
                 return 0xFF;
             } else {
-                uint8_t bitsWasted = oldDepth - newDepth;
+                uint8_t bitsWasted = static_cast<uint8_t>(oldDepth - newDepth);
                 uint16_t v = static_cast<uint16_t>(val);
-                v = (v + (1 << (bitsWasted - 1))) >> bitsWasted;
-                v = ::std::min<uint16_t>(::std::max<uint16_t>(0, v), (1 << newDepth) - 1);
+                v = static_cast<uint16_t>((v + (1 << (bitsWasted - 1))) >> bitsWasted);
+                v = ::std::min<uint16_t>(::std::max<uint16_t>(0, v),
+                                         static_cast<uint16_t>((1 << newDepth) - 1));
                 return static_cast<uint8_t>(v);
             }
         }
@@ -1191,18 +1192,18 @@ static uint32_t SelectPartition(int32_t seed, int32_t x, int32_t y, int32_t z,
     uint8_t seed11 = static_cast<uint8_t>((rnum >> 26) & 0xF);
     uint8_t seed12 = static_cast<uint8_t>(((rnum >> 30) | (rnum << 2)) & 0xF);
 
-    seed1 *= seed1;
-    seed2 *= seed2;
-    seed3 *= seed3;
-    seed4 *= seed4;
-    seed5 *= seed5;
-    seed6 *= seed6;
-    seed7 *= seed7;
-    seed8 *= seed8;
-    seed9 *= seed9;
-    seed10 *= seed10;
-    seed11 *= seed11;
-    seed12 *= seed12;
+    seed1 = static_cast<uint8_t>(seed1 * seed1);
+    seed2 = static_cast<uint8_t>(seed2 * seed2);
+    seed3 = static_cast<uint8_t>(seed3 * seed3);
+    seed4 = static_cast<uint8_t>(seed4 * seed4);
+    seed5 = static_cast<uint8_t>(seed5 * seed5);
+    seed6 = static_cast<uint8_t>(seed6 * seed6);
+    seed7 = static_cast<uint8_t>(seed7 * seed7);
+    seed8 = static_cast<uint8_t>(seed8 * seed8);
+    seed9 = static_cast<uint8_t>(seed9 * seed9);
+    seed10 = static_cast<uint8_t>(seed10 * seed10);
+    seed11 = static_cast<uint8_t>(seed11 * seed11);
+    seed12 = static_cast<uint8_t>(seed12 * seed12);
 
     int32_t sh1, sh2, sh3;
     if (seed & 1) {
@@ -1214,18 +1215,18 @@ static uint32_t SelectPartition(int32_t seed, int32_t x, int32_t y, int32_t z,
     }
     sh3 = (seed & 0x10) ? sh1 : sh2;
 
-    seed1 >>= sh1;
-    seed2 >>= sh2;
-    seed3 >>= sh1;
-    seed4 >>= sh2;
-    seed5 >>= sh1;
-    seed6 >>= sh2;
-    seed7 >>= sh1;
-    seed8 >>= sh2;
-    seed9 >>= sh3;
-    seed10 >>= sh3;
-    seed11 >>= sh3;
-    seed12 >>= sh3;
+    seed1 = static_cast<uint8_t>(seed1 >> sh1);
+    seed2 = static_cast<uint8_t>(seed2 >> sh2);
+    seed3 = static_cast<uint8_t>(seed3 >> sh1);
+    seed4 = static_cast<uint8_t>(seed4 >> sh2);
+    seed5 = static_cast<uint8_t>(seed5 >> sh1);
+    seed6 = static_cast<uint8_t>(seed6 >> sh2);
+    seed7 = static_cast<uint8_t>(seed7 >> sh1);
+    seed8 = static_cast<uint8_t>(seed8 >> sh2);
+    seed9 = static_cast<uint8_t>(seed9 >> sh3);
+    seed10 = static_cast<uint8_t>(seed10 >> sh3);
+    seed11 = static_cast<uint8_t>(seed11 >> sh3);
+    seed12 = static_cast<uint8_t>(seed12 >> sh3);
 
     int32_t a = seed1 * x + seed2 * y + seed11 * z + (rnum >> 14);
     int32_t b = seed3 * x + seed4 * y + seed12 * z + (rnum >> 10);
@@ -1558,7 +1559,9 @@ static void DecompressBlock(const uint8_t inBuf[16], const uint32_t blockWidth,
 
     // Make sure that higher non-texel bits are set to zero
     const uint32_t clearByteStart = (weightParams.GetPackedBitSize() >> 3) + 1;
-    texelWeightData[clearByteStart - 1] &= (1 << (weightParams.GetPackedBitSize() % 8)) - 1;
+    texelWeightData[clearByteStart - 1] =
+        texelWeightData[clearByteStart - 1] &
+        static_cast<uint8_t>((1 << (weightParams.GetPackedBitSize() % 8)) - 1);
     memset(texelWeightData + clearByteStart, 0, 16 - clearByteStart);
 
     std::vector<IntegerEncodedValue> texelWeightValues;
diff --git a/src/video_core/textures/texture.h b/src/video_core/textures/texture.h
index 27c8ce975..8e82c6748 100644
--- a/src/video_core/textures/texture.h
+++ b/src/video_core/textures/texture.h
@@ -342,13 +342,14 @@ struct TSCEntry {
     float GetLodBias() const {
         // Sign extend the 13-bit value.
         constexpr u32 mask = 1U << (13 - 1);
-        return static_cast<s32>((mip_lod_bias ^ mask) - mask) / 256.0f;
+        return static_cast<float>(static_cast<s32>((mip_lod_bias ^ mask) - mask)) / 256.0f;
     }
 
     std::array<float, 4> GetBorderColor() const {
         if (srgb_conversion) {
-            return {srgb_border_color_r / 255.0f, srgb_border_color_g / 255.0f,
-                    srgb_border_color_b / 255.0f, border_color[3]};
+            return {static_cast<float>(srgb_border_color_r) / 255.0f,
+                    static_cast<float>(srgb_border_color_g) / 255.0f,
+                    static_cast<float>(srgb_border_color_b) / 255.0f, border_color[3]};
         }
         return border_color;
     }
diff --git a/src/video_core/video_core.cpp b/src/video_core/video_core.cpp
index 60cda0ca3..8e947394c 100644
--- a/src/video_core/video_core.cpp
+++ b/src/video_core/video_core.cpp
@@ -28,7 +28,7 @@ std::unique_ptr<Tegra::GPU> CreateGPU(Core::System& system) {
 
 u16 GetResolutionScaleFactor(const RendererBase& renderer) {
     return static_cast<u16>(
-        Settings::values.resolution_factor
+        Settings::values.resolution_factor != 0
             ? Settings::values.resolution_factor
             : renderer.GetRenderWindow().GetFramebufferLayout().GetScalingRatio());
 }