renderer: decode R32_SFLOAT packed colour clears (#436)

* renderer: decode R32_SFLOAT packed colour clears

DecodePackedColorClear handled R32Uint, R32Sint and the 8/10-bit formats but fell through to the default false case for R32Sfloat. metadata_clear_supported then went false, the clear value was discarded, and the target kept its previous contents. Decode the packed word as float bits, matching DecodePackedDepthClear directly below, and reject non-finite patterns rather than materializing them.

* renderer: preserve all R32_SFLOAT clear bits

---------

Co-authored-by: nmzik <Nmzik@mail.ru>

.github/workflows/build.yml

@@ -208,34 +208,66 @@ jobs:
         run: |
           archive="$RUNNER_TEMP/MoltenVK-macos.tar"
           package="$RUNNER_TEMP/MoltenVK"
+          install_dir="_Build/macos/install"
+          app_dir="$install_dir/KytyPS5.app"
+          moltenvk="$install_dir/libMoltenVK.dylib"
           curl --fail --location --retry 3 \
             --output "$archive" \
             "https://github.com/KhronosGroup/MoltenVK/releases/download/$MOLTENVK_VERSION/MoltenVK-macos.tar"
           echo "$MOLTENVK_SHA256  $archive" | shasum -a 256 --check
           tar -xf "$archive" -C "$RUNNER_TEMP"
+          test -d "$app_dir"
           install -m 755 \
             "$package/MoltenVK/dynamic/dylib/macOS/libMoltenVK.dylib" \
-            _Build/macos/install/libMoltenVK.dylib
-          install -m 644 "$package/LICENSE" _Build/macos/install/LICENSE.MoltenVK
-          codesign --force --sign - --timestamp=none _Build/macos/install/libMoltenVK.dylib
+            "$moltenvk"
+          install -m 644 "$package/LICENSE" "$install_dir/LICENSE.MoltenVK"
+          codesign --force --sign - --timestamp=none "$moltenvk"
+          install -d -m 755 "$app_dir/Contents/Frameworks" "$app_dir/Contents/Resources"
+          install -m 755 "$moltenvk" "$app_dir/Contents/Frameworks/libMoltenVK.dylib"
+          install -m 644 "$package/LICENSE" "$app_dir/Contents/Resources/LICENSE.MoltenVK"
+          codesign --force --sign - --timestamp=none "$app_dir"
 
       - name: Verify artifacts
         shell: bash
         run: |
-          file _Build/macos/install/launcher
-          file _Build/macos/install/kyty_emulator
-          file _Build/macos/install/libMoltenVK.dylib
-          lipo _Build/macos/install/launcher -verify_arch x86_64
-          lipo _Build/macos/install/kyty_emulator -verify_arch x86_64
-          lipo _Build/macos/install/libMoltenVK.dylib -verify_arch x86_64
-          codesign --verify --strict _Build/macos/install/kyty_emulator
-          codesign --verify --strict _Build/macos/install/libMoltenVK.dylib
+          install_dir="_Build/macos/install"
+          app_dir="$install_dir/KytyPS5.app"
+          test -d "$app_dir"
+          for binary in \
+            "$install_dir/kyty_emulator" \
+            "$install_dir/libMoltenVK.dylib" \
+            "$app_dir/Contents/MacOS/KytyPS5" \
+            "$app_dir/Contents/MacOS/kyty_emulator" \
+            "$app_dir/Contents/Frameworks/libMoltenVK.dylib"; do
+            file "$binary"
+            lipo "$binary" -verify_arch x86_64
+          done
+          for binary in \
+            "$install_dir/kyty_emulator" \
+            "$install_dir/libMoltenVK.dylib" \
+            "$app_dir/Contents/MacOS/KytyPS5" \
+            "$app_dir/Contents/MacOS/kyty_emulator" \
+            "$app_dir/Contents/Frameworks/libMoltenVK.dylib"; do
+            codesign --verify --strict "$binary"
+          done
+          codesign --verify --deep --strict --verbose=2 "$app_dir"
+
+      - name: Package macOS artifacts
+        shell: bash
+        run: |
+          install_dir="_Build/macos/install"
+          archive="$RUNNER_TEMP/KytyPS5-macOS-x86_64.zip"
+          (
+            cd "$install_dir"
+            zip -qry "$archive" .
+          )
+          unzip -t "$archive"
 
       - name: Upload macOS artifacts
         uses: actions/upload-artifact@v4
         with:
           name: KytyPS5-macOS-x86_64
-          path: _Build/macos/install/**
+          path: ${{ runner.temp }}/KytyPS5-macOS-x86_64.zip
           if-no-files-found: error
 
   linux:
@@ -424,16 +456,9 @@ jobs:
         if: needs.macos.result == 'success'
         shell: bash
         run: |
-          macos_dir="artifacts/KytyPS5-macOS-x86_64"
-          test -d "$macos_dir"
-          chmod a+x \
-            "$macos_dir/launcher" \
-            "$macos_dir/kyty_emulator" \
-            "$macos_dir/libMoltenVK.dylib"
-          (
-            cd "$macos_dir"
-            zip -r "$GITHUB_WORKSPACE/$RELEASE_NAME-macOS-x86_64.zip" .
-          )
+          macos_archive="artifacts/KytyPS5-macOS-x86_64/KytyPS5-macOS-x86_64.zip"
+          test -f "$macos_archive"
+          cp "$macos_archive" "$RELEASE_NAME-macOS-x86_64.zip"
 
       - name: Package Linux build
         if: needs.linux.result == 'success'

.gitignore

@@ -5,3 +5,4 @@ build/
 _Build/vscode-clang/
 _Build/
 cmake-build-debug/
+_PipelineCache/

.vscode/launch.json

@@ -10,7 +10,14 @@
             "stopAtEntry": false,
             "cwd": "${workspaceFolder}\\_Build\\vscode-clang\\launcher",
             "environment": [],
-            "externalConsole": false
+            "externalConsole": false,
+            "linux": {
+                "type": "cppdbg",
+                "program": "${workspaceFolder}/_Build/vscode-clang/launcher/launcher",
+                "cwd": "${workspaceFolder}/_Build/vscode-clang/launcher",
+                "MIMode": "gdb",
+                "miDebuggerPath": "/usr/bin/gdb"
+            }
         },
         {
             "name": "Debug kyty_emulator",
@@ -25,13 +32,40 @@
             "stopAtEntry": false,
             "cwd": "${workspaceFolder}\\_Build\\vscode-clang",
             "environment": [],
-            "externalConsole": true
+            "externalConsole": true,
+            "linux": {
+                "type": "cppdbg",
+                "program": "${workspaceFolder}/_Build/vscode-clang/kyty_emulator",
+                "cwd": "${workspaceFolder}/_Build/vscode-clang",
+                "externalConsole": false,
+                "MIMode": "gdb",
+                "miDebuggerPath": "/usr/bin/gdb",
+                "setupCommands": [
+                    {
+                        "description": "Pass Kyty's internal signals to its handlers",
+                        "text": "handle SIG37 SIGSEGV SIGBUS SIGILL nostop noprint pass",
+                        "ignoreFailures": false
+                    }
+                ]
+            }
         },
         {
             "name": "Attach to running kyty_emulator",
             "type": "cppvsdbg",
             "request": "attach",
-            "processId": "${command:pickProcess}"
+            "processId": "${command:pickProcess}",
+            "linux": {
+                "type": "cppdbg",
+                "MIMode": "gdb",
+                "miDebuggerPath": "/usr/bin/gdb",
+                "setupCommands": [
+                    {
+                        "description": "Pass Kyty's internal signals to its handlers",
+                        "text": "handle SIG37 SIGSEGV SIGBUS SIGILL nostop noprint pass",
+                        "ignoreFailures": false
+                    }
+                ]
+            }
         }
     ]
 }

CMakeLists.txt

@@ -254,7 +254,7 @@ foreach(shader_name IN LISTS gpu_tiler_shader_names)
 	add_custom_command(
 		OUTPUT "${shader_header}"
 		COMMAND ${CMAKE_COMMAND} -E make_directory "${gpu_tiler_generated_dir}"
-		COMMAND "${KYTY_GLSLANG_VALIDATOR}" -V --target-env vulkan1.0 -Os
+		COMMAND "${KYTY_GLSLANG_VALIDATOR}" -V --target-env vulkan1.3 -Os
 		        "-I${gpu_tiler_shader_dir}" -o "${shader_spv}" "${shader_source}"
 		COMMAND ${CMAKE_COMMAND} -DINPUT=${shader_spv} -DOUTPUT=${shader_header}
 		        -DSYMBOL=${shader_symbol} -P "${KYTY_SOURCE_DIR}/embed_spirv.cmake"
@@ -263,6 +263,22 @@ foreach(shader_name IN LISTS gpu_tiler_shader_names)
 	)
 	list(APPEND gpu_tiler_shader_headers "${shader_header}")
 endforeach()
+
+set(fault_buffer_shader_source "${gpu_tiler_shader_dir}/fault_buffer_process.comp")
+set(fault_buffer_shader_spv "${gpu_tiler_generated_dir}/fault_buffer_process.spv")
+set(fault_buffer_shader_header "${gpu_tiler_generated_dir}/fault_buffer_process_spv.h")
+add_custom_command(
+	OUTPUT "${fault_buffer_shader_header}"
+	COMMAND ${CMAKE_COMMAND} -E make_directory "${gpu_tiler_generated_dir}"
+	COMMAND "${KYTY_GLSLANG_VALIDATOR}" -V --target-env vulkan1.3 -Os
+	        -DCACHING_PAGEBITS=14 -DMAX_PAGE_FAULTS=1024
+	        -o "${fault_buffer_shader_spv}" "${fault_buffer_shader_source}"
+	COMMAND ${CMAKE_COMMAND} -DINPUT=${fault_buffer_shader_spv} -DOUTPUT=${fault_buffer_shader_header}
+	        -DSYMBOL=FAULT_BUFFER_PROCESS_SPV -P "${KYTY_SOURCE_DIR}/embed_spirv.cmake"
+	DEPENDS "${fault_buffer_shader_source}"
+	VERBATIM
+)
+list(APPEND kyty_emulator_src "${fault_buffer_shader_header}")
 list(APPEND kyty_emulator_src ${gpu_tiler_shader_headers})
 
 set(gpu_blit_generated_dir "${PROJECT_BINARY_DIR}/gpu_blit_shaders")
@@ -278,7 +294,7 @@ foreach(shader_source IN LISTS gpu_blit_shader_sources)
 	add_custom_command(
 		OUTPUT "${shader_header}"
 		COMMAND ${CMAKE_COMMAND} -E make_directory "${gpu_blit_generated_dir}"
-		COMMAND "${KYTY_GLSLANG_VALIDATOR}" -V --target-env vulkan1.0 -Os
+		COMMAND "${KYTY_GLSLANG_VALIDATOR}" -V --target-env vulkan1.3 -Os
 		        "-I${gpu_tiler_shader_dir}" -o "${shader_spv}" "${shader_source}"
 		COMMAND ${CMAKE_COMMAND} -DINPUT=${shader_spv} -DOUTPUT=${shader_header}
 		        -DSYMBOL=${shader_symbol} -P "${KYTY_SOURCE_DIR}/embed_spirv.cmake"
@@ -430,6 +446,7 @@ add_executable(resource_tracking_tests EXCLUDE_FROM_ALL
 	"${KYTY_TESTS_DIR}/ResourceTrackingTests.cpp"
 	"${KYTY_SOURCE_DIR}/graphics/guest_gpu/gpu_format.cpp"
 	"${KYTY_SOURCE_DIR}/graphics/host_gpu/hostMemory.cpp"
+	"${KYTY_SOURCE_DIR}/graphics/shader/shaderPixelParameter.cpp"
 	"${KYTY_SOURCE_DIR}/graphics/shader/recompiler/ir/passes/SrtWalker.cpp"
 	"${KYTY_SOURCE_DIR}/graphics/shader/recompiler/ir/passes/ResourceTracking.cpp"
 	"${KYTY_SOURCE_DIR}/graphics/shader/recompiler/ir/passes/ResourceMaterialization.cpp"
@@ -483,7 +500,7 @@ set(gpu_test_ms_depth_header
 add_custom_command(
 	OUTPUT "${gpu_test_ms_depth_header}"
 	COMMAND ${CMAKE_COMMAND} -E make_directory "${gpu_test_generated_dir}"
-	COMMAND "${KYTY_GLSLANG_VALIDATOR}" -V --target-env vulkan1.0 -Os
+	COMMAND "${KYTY_GLSLANG_VALIDATOR}" -V --target-env vulkan1.3 -Os
 	        -o "${gpu_test_ms_depth_spv}" "${gpu_test_ms_depth_source}"
 	COMMAND ${CMAKE_COMMAND} -DINPUT=${gpu_test_ms_depth_spv}
 	        -DOUTPUT=${gpu_test_ms_depth_header} -DSYMBOL=GPU_TEST_MS_DEPTH_SPV
@@ -507,6 +524,16 @@ if(NOT KYTY_CLANG_CL)
 endif()
 
 if(BUILD_TESTING)
+	add_test(NAME emulator_user_name_cli
+	         COMMAND $<TARGET_FILE:kyty_emulator> --help --user-name 1234567890123456)
+	add_test(NAME emulator_user_name_too_long
+	         COMMAND $<TARGET_FILE:kyty_emulator> --help --user-name 12345678901234567)
+	set_tests_properties(emulator_user_name_too_long PROPERTIES WILL_FAIL TRUE)
+	add_test(NAME emulator_present_mode_cli
+	         COMMAND $<TARGET_FILE:kyty_emulator> --help --present-mode Immediate)
+	add_test(NAME emulator_present_mode_invalid
+	         COMMAND $<TARGET_FILE:kyty_emulator> --help --present-mode Invalid)
+	set_tests_properties(emulator_present_mode_invalid PROPERTIES WILL_FAIL TRUE)
 	add_test(NAME ime_dialog COMMAND $<TARGET_FILE:ime_dialog_tests>)
 	add_test(NAME shader_cfg COMMAND $<TARGET_FILE:shader_cfg_tests>)
 	add_test(NAME scalar_provenance COMMAND $<TARGET_FILE:scalar_provenance_tests>)

README.md

@@ -220,17 +220,23 @@ cmake --install _Build/macos --prefix _Build/macos/install
 ```
 
 The build re-signs `kyty_emulator` with the JIT entitlements it needs to execute translated
-guest code; no manual signing step is required.
+guest code; no manual signing step is required. When the launcher is built, the install
+also produces `_Build/macos/install/KytyPS5.app` — double-click to launch the GUI.
+A flat `kyty_emulator` is kept for CLI usage.
 
 Vulkan comes from MoltenVK. Download `MoltenVK-macos.tar` from the
 [MoltenVK releases](https://github.com/KhronosGroup/MoltenVK/releases), then copy
-`MoltenVK/dynamic/dylib/macOS/libMoltenVK.dylib` next to `kyty_emulator` and ad-hoc sign it:
+`MoltenVK/dynamic/dylib/macOS/libMoltenVK.dylib` next to the flat `kyty_emulator`
+(and, for the bundle, into `KytyPS5.app/Contents/Frameworks/`) and ad-hoc sign it:
 
 ```bash
 codesign --force --sign - _Build/macos/install/libMoltenVK.dylib
+# For the bundle (if present):
+codesign --force --sign - _Build/macos/install/KytyPS5.app/Contents/Frameworks/libMoltenVK.dylib
+codesign --force --sign - _Build/macos/install/KytyPS5.app
 ```
 
-Release archives already include a signed `libMoltenVK.dylib`.
+Release archives already include a signed `libMoltenVK.dylib` (both flat and inside the bundle).
 
 ### Regression tests
 
@@ -274,6 +280,10 @@ To use the graphical launcher:
 ./_Build/linux/install/launcher
 ```
 
+```bash
+open _Build/macos/install/KytyPS5.app  # or double-click in Finder
+```
+
 On first launch, add one or more game folders in the global settings. The launcher searches those
 folders recursively for game directories containing `eboot.bin`. Select a detected game and run it
 from the game list.
@@ -288,12 +298,17 @@ The emulator can also be started directly with a legally obtained game directory
 ./_Build/linux/install/kyty_emulator --game "/games/ExampleGame"
 ```
 
-On macOS, point SDL at the MoltenVK library explicitly; the hardened runtime prevents it from
-being picked up from the executable's directory:
+On macOS, the adjacent flat or app-bundled `libMoltenVK.dylib` is found automatically; no
+environment variable is required:
 
 ```bash
-cd _Build/macos/install
-SDL_VULKAN_LIBRARY="$PWD/libMoltenVK.dylib" ./kyty_emulator --game "/games/ExampleGame"
+./_Build/macos/install/kyty_emulator --game "/games/ExampleGame"
+```
+
+To override the Vulkan loader, set `SDL_VULKAN_LIBRARY`:
+
+```bash
+SDL_VULKAN_LIBRARY=/path/to/libMoltenVK.dylib ./kyty_emulator --game "/games/ExampleGame"
 ```
 
 Run `kyty_emulator --help` to see the available graphics, logging, validation, profiling, and

src/common/emulatorConfig.cpp

@@ -21,6 +21,8 @@ void Shutdown() {
 
 void Load(const ConfigOptions& cfg) {
 	EXIT_IF(g_config == nullptr);
+	EXIT_IF(cfg.user_name.empty() || cfg.user_name.size() > MAX_USER_NAME_LENGTH);
+	EXIT_IF(!IsConfiguredUserIdValid(cfg.user_id));
 
 	*g_config = cfg;
 }
@@ -33,6 +35,22 @@ uint32_t GetScreenHeight() {
 	return g_config->screen_height;
 }
 
+const std::string& GetUserName() {
+	return g_config->user_name;
+}
+
+int32_t GetUserId() {
+	return g_config->user_id;
+}
+
+PresentMode GetPresentMode() {
+	return g_config->present_mode;
+}
+
+int32_t GetGpuIndex() {
+	return g_config->gpu_index;
+}
+
 bool FullscreenEnabled() {
 	return g_config->fullscreen_enabled;
 }

src/common/emulatorConfig.h

@@ -3,6 +3,7 @@
 
 #include "common/common.h"
 
+#include <cstddef>
 #include <filesystem>
 #include <string>
 #include <vector>
@@ -26,14 +27,28 @@ enum class ProfilerDirection { None, Network };
 
 enum class OutputDirection { Silent, Console, File };
 
+enum class PresentMode { Fifo, Mailbox, Immediate };
+
 using Keymap = std::vector<std::string>;
 
 constexpr uint32_t DEFAULT_CONSOLE_LANGUAGE = 1;
 constexpr uint32_t MAX_CONSOLE_LANGUAGE     = 29;
+constexpr std::size_t MAX_USER_NAME_LENGTH = 16;
+constexpr int32_t DEFAULT_USER_ID           = 1000;
+
+constexpr bool IsConfiguredUserIdValid(int32_t user_id) {
+	constexpr int32_t USER_ID_EVERYONE = 0xfe;
+	constexpr int32_t USER_ID_SYSTEM   = 0xff;
+	return user_id >= 0 && user_id != USER_ID_EVERYONE && user_id != USER_ID_SYSTEM;
+}
 
 struct ConfigOptions {
 	uint32_t               screen_width                = 1280;
 	uint32_t               screen_height               = 720;
+	std::string            user_name                   = "Kyty";
+	int32_t                user_id                     = DEFAULT_USER_ID;
+	PresentMode            present_mode                = PresentMode::Fifo;
+	int32_t                gpu_index                   = -1;
 	bool                   fullscreen_enabled          = false;
 	uint32_t               vblank_frequency            = 60;
 	uint32_t               console_language            = DEFAULT_CONSOLE_LANGUAGE;
@@ -45,7 +60,7 @@ struct ConfigOptions {
 	bool                   command_buffer_dump_enabled = false;
 	std::filesystem::path  command_buffer_dump_folder  = "_Buffers";
 	bool                   graphics_debug_dump_enabled = false;
-	OutputDirection        printf_direction            = OutputDirection::Console;
+	OutputDirection        printf_direction            = OutputDirection::Silent;
 	std::filesystem::path  printf_output_file          = "_kyty.txt";
 	ProfilerDirection      profiler_direction          = ProfilerDirection::None;
 	bool                   spirv_debug_printf_enabled  = false;
@@ -63,6 +78,10 @@ void Load(const ConfigOptions& cfg);
 
 uint32_t GetScreenWidth();
 uint32_t GetScreenHeight();
+const std::string& GetUserName();
+int32_t  GetUserId();
+PresentMode GetPresentMode();
+int32_t GetGpuIndex();
 bool     FullscreenEnabled();
 uint32_t GetVblankFrequency();
 uint32_t GetConsoleLanguage();

src/common/file.cpp

@@ -25,9 +25,6 @@
 #ifdef CopyFile
 #undef CopyFile
 #endif
-#ifdef MoveFile
-#undef MoveFile
-#endif
 #endif
 
 namespace Common {
@@ -544,14 +541,14 @@ bool File::CopyFile(const std::filesystem::path& src,
 	return SysFileCopyFile(src, dst);
 }
 
-bool File::MoveFile(const std::filesystem::path& src,
-                    const std::filesystem::path& dst) // @suppress("Member declaration not found")
+bool File::RenameFile(const std::filesystem::path& src,
+                      const std::filesystem::path& dst) // @suppress("Member declaration not found")
 {
 	if (IsFileExisting(dst)) {
 		DeleteFile(dst); // @suppress("Invalid arguments")
 	}
 
-	return SysFileMoveFile(src, dst);
+	return SysFileRenameFile(src, dst);
 }
 
 void File::RemoveReadonly(const std::filesystem::path& name) {

src/common/file.h

@@ -20,9 +20,6 @@
 #ifdef CopyFile
 #undef CopyFile
 #endif
-#ifdef MoveFile
-#undef MoveFile
-#endif
 #endif
 
 namespace Common {
@@ -107,7 +104,7 @@ public:
 	static std::vector<FindInfo> FindFiles(const std::filesystem::path& path);
 	static std::vector<DirEntry> GetDirEntries(const std::filesystem::path& path);
 	static bool CopyFile(const std::filesystem::path& src, const std::filesystem::path& dst);
-	static bool MoveFile(const std::filesystem::path& src, const std::filesystem::path& dst);
+	static bool RenameFile(const std::filesystem::path& src, const std::filesystem::path& dst);
 	static void RemoveReadonly(const std::filesystem::path& name);
 
 	KYTY_CLASS_NO_COPY(File);

src/common/platform/sysFileIO.h

@@ -80,7 +80,7 @@ void SysFileFindFiles(const std::filesystem::path& path, std::vector<sys_file_fi
 // NOLINTNEXTLINE(google-runtime-references)
 void SysFileGetDents(const std::filesystem::path& path, std::vector<sys_dir_entry_t>& out);
 bool SysFileCopyFile(const std::filesystem::path& src, const std::filesystem::path& dst);
-bool SysFileMoveFile(const std::filesystem::path& src, const std::filesystem::path& dst);
+bool SysFileRenameFile(const std::filesystem::path& src, const std::filesystem::path& dst);
 void SysFileRemoveReadonly(const std::filesystem::path& name);
 
 #endif /* KYTY_COMMON_PLATFORM_SYSFILEIO_H_ */

src/common/platform/sysLinuxFileIO.cpp

@@ -693,7 +693,7 @@ bool SysFileCopyFile(const std::filesystem::path& src, const std::filesystem::pa
 	       !error;
 }
 
-bool SysFileMoveFile(const std::filesystem::path& src, const std::filesystem::path& dst) {
+bool SysFileRenameFile(const std::filesystem::path& src, const std::filesystem::path& dst) {
 	auto real_src = get_internal_name(src);
 	auto real_dst = get_internal_name(dst);
 

src/common/platform/sysWindowsFileIO.cpp

@@ -565,7 +565,7 @@ bool SysFileCopyFile(const std::filesystem::path& src, const std::filesystem::pa
 	return CopyFileW(src_wide.c_str(), dst_wide.c_str(), FALSE) != 0;
 }
 
-bool SysFileMoveFile(const std::filesystem::path& src, const std::filesystem::path& dst) {
+bool SysFileRenameFile(const std::filesystem::path& src, const std::filesystem::path& dst) {
 	auto src_wide = src.wstring();
 	auto dst_wide = dst.wstring();
 	return MoveFileW(src_wide.c_str(), dst_wide.c_str()) != 0;

src/common/slotVector.h

@@ -0,0 +1,132 @@
+#ifndef EMULATOR_SRC_COMMON_SLOTVECTOR_H_
+#define EMULATOR_SRC_COMMON_SLOTVECTOR_H_
+
+#include "common/assert.h"
+#include "common/abi.h"
+
+#include <compare>
+#include <cstdint>
+#include <deque>
+#include <functional>
+#include <limits>
+#include <optional>
+#include <utility>
+#include <vector>
+
+namespace Common {
+
+struct SlotId {
+	static constexpr uint32_t INVALID_INDEX = std::numeric_limits<uint32_t>::max();
+
+	constexpr SlotId() noexcept = default;
+	constexpr SlotId(uint32_t value) noexcept: index(value), generation(1) {}
+	constexpr SlotId(uint32_t value, uint32_t slot_generation) noexcept
+	    : index(value), generation(slot_generation) {}
+
+	[[nodiscard]] constexpr explicit operator bool() const noexcept { return index != INVALID_INDEX; }
+	constexpr auto operator<=>(const SlotId&) const noexcept = default;
+
+	uint32_t index = INVALID_INDEX;
+	uint32_t generation = 0;
+};
+
+// Stable-address slot storage for cache resources that are intentionally non-movable.
+template <typename T>
+class SlotVector {
+public:
+	SlotVector() = default;
+	KYTY_CLASS_NO_COPY(SlotVector);
+
+	[[nodiscard]] T& operator[](SlotId id) noexcept {
+		EXIT_IF(!is_allocated(id));
+		return *m_values[id.index].value;
+	}
+
+	[[nodiscard]] const T& operator[](SlotId id) const noexcept {
+		EXIT_IF(!is_allocated(id));
+		return *m_values[id.index].value;
+	}
+
+	[[nodiscard]] T* try_get(SlotId id) noexcept {
+		return is_allocated(id) ? &*m_values[id.index].value : nullptr;
+	}
+
+	[[nodiscard]] const T* try_get(SlotId id) const noexcept {
+		return is_allocated(id) ? &*m_values[id.index].value : nullptr;
+	}
+
+	[[nodiscard]] bool is_allocated(SlotId id) const noexcept {
+		return id && id.index < m_values.size() &&
+		       m_values[id.index].generation == id.generation && m_values[id.index].value.has_value();
+	}
+
+	template <typename... Args>
+	[[nodiscard]] SlotId insert(Args&&... args) {
+		uint32_t index = 0;
+		if (m_free_list.empty()) {
+			index = static_cast<uint32_t>(m_values.size());
+			m_values.emplace_back();
+			m_values.back().value.emplace(std::forward<Args>(args)...);
+		} else {
+			index = m_free_list.back();
+			m_free_list.pop_back();
+			EXIT_IF(m_values[index].value.has_value());
+			m_values[index].value.emplace(std::forward<Args>(args)...);
+		}
+		++m_size;
+		return SlotId {index, m_values[index].generation};
+	}
+
+	void erase(SlotId id) noexcept {
+		EXIT_IF(!is_allocated(id));
+		auto& slot = m_values[id.index];
+		slot.value.reset();
+		if (++slot.generation == 0) {
+			slot.generation = 1;
+		}
+		m_free_list.push_back(id.index);
+		--m_size;
+	}
+
+	[[nodiscard]] size_t size() const noexcept { return m_size; }
+	[[nodiscard]] size_t capacity() const noexcept { return m_values.size(); }
+
+	template <typename F>
+	void ForEach(F&& fn) {
+		for (uint32_t index = 0; index < m_values.size(); ++index) {
+			if (m_values[index].value) {
+				fn(SlotId {index, m_values[index].generation}, *m_values[index].value);
+			}
+		}
+	}
+
+	template <typename F>
+	void ForEach(F&& fn) const {
+		for (uint32_t index = 0; index < m_values.size(); ++index) {
+			if (m_values[index].value) {
+				fn(SlotId {index, m_values[index].generation}, *m_values[index].value);
+			}
+		}
+	}
+
+private:
+	struct Slot {
+		std::optional<T> value;
+		uint32_t         generation = 1;
+	};
+
+	std::deque<Slot>             m_values;
+	std::vector<uint32_t>        m_free_list;
+	size_t                       m_size = 0;
+};
+
+} // namespace Common
+
+template <>
+struct std::hash<Common::SlotId> {
+	[[nodiscard]] size_t operator()(Common::SlotId id) const noexcept {
+		return std::hash<uint64_t> {}((static_cast<uint64_t>(id.generation) << 32u) | id.index);
+	}
+};
+
+#endif // EMULATOR_SRC_COMMON_SLOTVECTOR_H_

src/emulator.cpp

@@ -15,6 +15,7 @@
 #include "kernel/fileSystem.h"
 #include "kernel/memory.h"
 #include "kernel/pthread.h"
+#include "kytyGitVersion.h"
 #include "libs/agc.h"
 #include "libs/audio.h"
 #include "libs/controller.h"
@@ -26,13 +27,31 @@
 
 #include <cstdlib>
 #include <filesystem>
+#include <thread>
 
 namespace Emulator {
 
 static void PrintSystemInfo() {
-	Common::SystemInfo info = Common::GetSystemInfo();
+	const Common::SystemInfo info = Common::GetSystemInfo();
 
-	LOGF("ProcessorName = %s\n", info.ProcessorName.c_str());
+#if defined(__APPLE__)
+	static constexpr auto platform_name = "macOS";
+#elif KYTY_PLATFORM == KYTY_PLATFORM_WINDOWS
+	static constexpr auto platform_name = "Windows";
+#elif KYTY_PLATFORM == KYTY_PLATFORM_LINUX
+	static constexpr auto platform_name = "Linux";
+#else
+	static constexpr auto platform_name = "Unknown";
+#endif
+
+	LOGF("Build\n"
+	     "  version: %s\n\n"
+	     "Host\n"
+	     "  os:      %s\n"
+	     "  cpu:     %s\n"
+	     "  threads: %u\n\n",
+	     KYTY_BUILD_LABEL, platform_name, info.ProcessorName.c_str(),
+	     std::thread::hardware_concurrency());
 }
 
 static void KytyClose() {
@@ -181,6 +200,10 @@ void Run(const RunOptions& options) {
 	int ok = atexit(KytyClose);
 	EXIT_NOT_IMPLEMENTED(ok != 0);
 
+	// Guest threads are still running, so skip KytyClose() and only flush emergency state.
+	ok = at_quick_exit(Common::Subsystems::EmergencyShutdownActive);
+	EXIT_NOT_IMPLEMENTED(ok != 0);
+
 	Libs::LibKernel::FileSystem::Mount(options.app0_dir, "/app0");
 	Libs::LibKernel::FileSystem::Mount(options.app0_dir, "/hostapp");
 

src/generate_version.cmake

@@ -1,5 +1,6 @@
 set(KYTY_GIT_VERSION "unknown")
 set(KYTY_GIT_HASH "unknown")
+set(KYTY_GIT_REVISION "unknown")
 if(GIT_EXECUTABLE)
 	execute_process(
 		COMMAND "${GIT_EXECUTABLE}" describe --tags --always --dirty
@@ -14,16 +15,18 @@ if(GIT_EXECUTABLE)
 	endif()
 
 	execute_process(
-		COMMAND "${GIT_EXECUTABLE}" rev-parse --short=7 HEAD
+		COMMAND "${GIT_EXECUTABLE}" rev-parse HEAD
 		WORKING_DIRECTORY "${GIT_WORKING_DIRECTORY}"
-		OUTPUT_VARIABLE KYTY_GIT_HASH
+		OUTPUT_VARIABLE KYTY_GIT_REVISION
 		OUTPUT_STRIP_TRAILING_WHITESPACE
 		RESULT_VARIABLE GIT_HASH_RESULT
 		ERROR_QUIET
 	)
 	if(NOT GIT_HASH_RESULT EQUAL 0)
 		set(KYTY_GIT_HASH "unknown")
+		set(KYTY_GIT_REVISION "unknown")
 	else()
+		string(SUBSTRING "${KYTY_GIT_REVISION}" 0 7 KYTY_GIT_HASH)
 		execute_process(
 			COMMAND "${GIT_EXECUTABLE}" diff-index --quiet HEAD --
 			WORKING_DIRECTORY "${GIT_WORKING_DIRECTORY}"

src/graphics/guest_gpu/command_processor/commandProcessor.h

@@ -7,6 +7,7 @@
 #include "graphics/host_gpu/renderer/renderContext.h"
 
 #include <cstdint>
+#include <span>
 #include <vector>
 
 namespace Libs::Graphics {
@@ -30,10 +31,9 @@ private:
 	friend class CommandProcessor;
 
 	struct BufferCursor {
-		uint32_t* next_packet         = nullptr;
-		uint32_t  remaining_dw        = 0;
-		uint32_t  total_dw            = 0;
-		uint32_t  deferred_advance_dw = 0;
+		std::span<const uint32_t> commands;
+		uint32_t                  offset_dw           = 0;
+		uint32_t                  deferred_advance_dw = 0;
 	};
 
 	std::vector<BufferCursor> m_buffer_stack;
@@ -50,7 +50,8 @@ public:
 		int64_t flip_arg  = 0;
 	};
 
-	explicit CommandProcessor(RenderContext& renderer): m_renderer(renderer) {}
+	CommandProcessor(RenderContext& renderer, int interrupt_event_id)
+	    : m_renderer(renderer), m_interrupt_event_id(interrupt_event_id) {}
 	~CommandProcessor() = default;
 
 	KYTY_CLASS_NO_COPY(CommandProcessor);
@@ -103,7 +104,7 @@ public:
 	                    uint32_t mode);
 	void DispatchIndirect(uint32_t data_offset, uint32_t mode);
 	void WaitFlipDone(uint32_t video_out_handle, uint32_t display_buffer_index);
-	void TriggerEvent(uint32_t event_type, uint32_t event_index);
+	void TriggerEvent(uint32_t event_type, uint32_t event_index, uint64_t event_address = 0);
 
 	void SetUserDataMarker(HW::UserSgprType type) { m_user_data_marker = type; }
 	[[nodiscard]] HW::UserSgprType GetUserDataMarker() const { return m_user_data_marker; }
@@ -131,13 +132,14 @@ public:
 	                    const volatile void* address, uint32_t count_in_dwords);
 	[[nodiscard]] bool ShouldSkipPredicatedPackets() const { return m_predicate_skip; }
 
-	Pm4ProcessResult Process(Pm4Execution& execution, uint32_t* buffer, uint32_t size_dw);
-	void             ProcessIndirectBuffer(uint32_t* buffer, uint32_t size_dw);
+	Pm4ProcessResult Process(Pm4Execution& execution, std::span<const uint32_t> commands);
+	void             ProcessIndirectBuffer(std::span<const uint32_t> commands);
 
 	void SetFlip(const FlipInfo& flip) { m_flip = flip; }
 
 	[[nodiscard]] uint64_t GetSubmitId() const { return m_submit_id; }
 	void                   SetSubmitId(uint64_t submit_id) { m_submit_id = submit_id; }
+	[[nodiscard]] bool     IsAsyncComputeQueue() const { return m_interrupt_event_id >= 0x20; }
 
 private:
 	template <typename T>
@@ -151,10 +153,10 @@ private:
 	                          uint32_t render_target_slice_offset, uint32_t first_vertex,
 	                          uint32_t first_instance);
 
-	CommandScheduler&    GetScheduler() const { return m_renderer.GetCommandScheduler(); }
-	RenderCommandBuffer& CurrentBuffer() { return GetScheduler().Current(); }
-	void                 CheckBuffer() const { GetScheduler().CheckActive(); }
-	GpuResourceManager&  GetGpuResources() const { return m_renderer.GetGpuResources(); }
+	CommandScheduler&   GetScheduler() const { return m_renderer.GetCommandScheduler(); }
+	CommandBuffer&      CurrentBuffer() { return GetScheduler().Current(); }
+	void                CheckBuffer() const { GetScheduler().CheckActive(); }
+	GpuResourceManager& GetGpuResources() const { return m_renderer.GetGpuResources(); }
 
 	RenderContext&   m_renderer;
 	HW::Context      m_ctx;
@@ -177,9 +179,11 @@ private:
 
 	uint32_t m_const_ram[0x3000] = {0};
 
-	FlipInfo m_flip;
-	uint64_t m_submit_id      = 0;
-	bool     m_predicate_skip = false;
+	FlipInfo  m_flip;
+	const int m_interrupt_event_id;
+	uint64_t  m_submit_id                   = 0;
+	uint64_t  m_synthetic_occlusion_counter = 0;
+	bool      m_predicate_skip              = false;
 };
 
 } // namespace Libs::Graphics

src/graphics/guest_gpu/command_processor/pm4Dispatch.cpp

@@ -23,9 +23,8 @@ constexpr auto MakeContextDispatchTable() {
 	g_hw_ctx_func[Pm4::DB_STENCIL_CLEAR]             = HwCtxSetStencilClear;
 	g_hw_ctx_func[Pm4::DB_DEPTH_CLEAR]               = HwCtxSetDepthClear;
 	g_hw_ctx_func[Pm4::PA_SC_SCREEN_SCISSOR_TL]      = HwCtxSetScreenScissor;
-	g_hw_ctx_func[Pm4::DB_Z_INFO]                    = HwCtxSetDepthRenderTarget;
 	g_hw_ctx_func[Pm4::DB_STENCIL_INFO]              = HwCtxSetStencilInfo;
-	g_hw_ctx_func[Pm4::DB_HTILE_SURFACE]             = HwCtxSetDepthHtileSurface;
+	g_hw_ctx_func[Pm4::DB_SHADING_RATE_ENCODING]     = HwCtxSetDepthShadingRateEncoding;
 	g_hw_ctx_func[Pm4::PA_SU_HARDWARE_SCREEN_OFFSET] = HwCtxSetHardwareScreenOffset;
 	g_hw_ctx_func[Pm4::PA_SC_WINDOW_OFFSET]          = HwCtxSetWindowOffset;
 	g_hw_ctx_func[Pm4::PA_SC_WINDOW_SCISSOR_TL]      = HwCtxSetWindowScissor;
@@ -75,7 +74,6 @@ constexpr auto MakeContextDispatchTable() {
 	g_hw_ctx_func[Pm4::PA_SC_BINNER_CNTL_0]                   = HwCtxSetPaScExtendedControl;
 	g_hw_ctx_func[Pm4::PA_SC_BINNER_CNTL_1]                   = HwCtxSetPaScExtendedControl;
 	g_hw_ctx_func[Pm4::PA_SC_CONSERVATIVE_RASTERIZATION_CNTL] = HwCtxSetPaScExtendedControl;
-	g_hw_ctx_func[Pm4::PA_SC_NGG_MODE_CNTL]                   = HwCtxSetPaScExtendedControl;
 	g_hw_ctx_func[Pm4::DB_ALPHA_TO_MASK]                      = HwCtxSetAlphaToMask;
 	g_hw_ctx_func[Pm4::VGT_DRAW_PAYLOAD_CNTL]                 = HwCtxSetDrawPayloadControl;
 	g_hw_ctx_func[Pm4::VGT_PRIMITIVEID_RESET]                 = HwCtxSetPrimitiveIdReset;
@@ -83,7 +81,6 @@ constexpr auto MakeContextDispatchTable() {
 	g_hw_ctx_func[Pm4::VGT_SHADER_STAGES_EN]                  = HwCtxSetShaderStages;
 
 	for (uint32_t slot = 0; slot < 8; slot++) {
-		g_hw_ctx_func[Pm4::CB_COLOR0_BASE + slot * 15] = HwCtxSetRenderTarget;
 		g_hw_ctx_func[Pm4::CB_COLOR0_INFO + slot * 15] = HwCtxSetColorInfo;
 
 		g_hw_ctx_func[Pm4::CB_BLEND0_CONTROL + slot * 1] = HwCtxSetBlendControl;
@@ -116,25 +113,21 @@ constexpr auto MakeShaderDispatchTable() {
 	}
 
 	for (uint32_t slot = 0; slot < 4; slot++) {
-		g_hw_sh_func[Pm4::SPI_SHADER_USER_ACCUM_PS_0 + slot] = HwShSetPsUserAccumSgpr;
+		g_hw_sh_func[Pm4::SPI_SHADER_USER_ACCUM_PS_0 + slot] = HwShIgnoreUserAccumulator;
 	}
 
 	for (uint32_t slot = 0; slot < 4; slot++) {
-		g_hw_sh_func[Pm4::SPI_SHADER_USER_ACCUM_ESGS_0 + slot] = HwShSetGsUserAccumSgpr;
+		g_hw_sh_func[Pm4::SPI_SHADER_USER_ACCUM_ESGS_0 + slot] = HwShIgnoreUserAccumulator;
 	}
 	for (uint32_t slot = 0; slot < 4; slot++) {
-		g_hw_sh_func[Pm4::SPI_SHADER_USER_ACCUM_LSHS_0 + slot] = HwShSetHsUserAccumSgpr;
+		g_hw_sh_func[Pm4::SPI_SHADER_USER_ACCUM_LSHS_0 + slot] = HwShIgnoreUserAccumulator;
 	}
 
-	for (uint32_t slot = 0; Pm4::COMPUTE_USER_ACCUM_0 + slot < Pm4::COMPUTE_USER_DATA_0; slot++) {
-		g_hw_sh_func[Pm4::COMPUTE_USER_ACCUM_0 + slot] = HwShSetCsUserAccumSgpr;
+	for (uint32_t slot = 0; slot < 4; slot++) {
+		g_hw_sh_func[Pm4::COMPUTE_USER_ACCUM_0 + slot] = HwShIgnoreUserAccumulator;
 	}
 	g_hw_sh_func[Pm4::COMPUTE_PGM_LO]                  = HwShSetCsRegisters;
 	g_hw_sh_func[Pm4::COMPUTE_PGM_HI]                  = HwShSetCsRegisters;
-	g_hw_sh_func[Pm4::COMPUTE_TBA_LO]                  = HwShSetCsRegisters;
-	g_hw_sh_func[Pm4::COMPUTE_TBA_HI]                  = HwShSetCsRegisters;
-	g_hw_sh_func[Pm4::COMPUTE_TMA_LO]                  = HwShSetCsRegisters;
-	g_hw_sh_func[Pm4::COMPUTE_TMA_HI]                  = HwShSetCsRegisters;
 	g_hw_sh_func[Pm4::COMPUTE_PGM_RSRC1]               = HwShSetCsRegisters;
 	g_hw_sh_func[Pm4::COMPUTE_PGM_RSRC2]               = HwShSetCsRegisters;
 	g_hw_sh_func[Pm4::COMPUTE_START_X]                 = HwShSetCsRegisters;
@@ -144,30 +137,16 @@ constexpr auto MakeShaderDispatchTable() {
 	g_hw_sh_func[Pm4::COMPUTE_NUM_THREAD_Y]            = HwShSetCsRegisters;
 	g_hw_sh_func[Pm4::COMPUTE_NUM_THREAD_Z]            = HwShSetCsRegisters;
 	g_hw_sh_func[Pm4::COMPUTE_RESOURCE_LIMITS]         = HwShSetCsRegisters;
-	g_hw_sh_func[Pm4::COMPUTE_DESTINATION_EN_SE0]      = HwShSetCsRegisters;
-	g_hw_sh_func[Pm4::COMPUTE_DESTINATION_EN_SE1]      = HwShSetCsRegisters;
 	g_hw_sh_func[Pm4::COMPUTE_TMPRING_SIZE]            = HwShSetCsRegisters;
-	g_hw_sh_func[Pm4::COMPUTE_DESTINATION_EN_SE2]      = HwShSetCsRegisters;
-	g_hw_sh_func[Pm4::COMPUTE_DESTINATION_EN_SE3]      = HwShSetCsRegisters;
 	g_hw_sh_func[Pm4::COMPUTE_PGM_RSRC3]               = HwShSetCsRegisters;
-	g_hw_sh_func[Pm4::COMPUTE_SHADER_CHKSUM]           = HwShSetCsRegisters;
-	g_hw_sh_func[Pm4::COMPUTE_DISPATCH_TUNNEL]         = HwShSetCsRegisters;
-	g_hw_sh_func[Pm4::SPI_SHADER_TBA_LO_PS]            = HwShIgnoreRegisters;
-	g_hw_sh_func[Pm4::SPI_SHADER_PGM_RSRC3_PS]         = HwShIgnoreRegisters;
-	g_hw_sh_func[Pm4::SPI_SHADER_PGM_RSRC4_PS]         = HwShIgnoreRegisters;
-	g_hw_sh_func[Pm4::SPI_SHADER_TMA_LO_PS]            = HwShIgnoreRegisters;
-	g_hw_sh_func[Pm4::SPI_SHADER_TMA_HI_PS]            = HwShIgnoreRegisters;
-	g_hw_sh_func[Pm4::SPI_SHADER_REQ_CTRL_PS]          = HwShIgnoreRegisters;
-	g_hw_sh_func[Pm4::SPI_SHADER_PGM_RSRC3_GS]         = HwShIgnoreRegisters;
-	g_hw_sh_func[Pm4::SPI_SHADER_PGM_RSRC4_GS]         = HwShIgnoreRegisters;
+	g_hw_sh_func[Pm4::COMPUTE_PACE_ID]                 = HwShSetCsRegisters;
+	g_hw_sh_func[Pm4::SPI_GRAPHICS_SHADER_CONTROL_PS]  = HwShIgnoreRegisters;
+	g_hw_sh_func[Pm4::SPI_GRAPHICS_SHADER_CONTROL_GS]  = HwShIgnoreRegisters;
 	g_hw_sh_func[Pm4::SPI_SHADER_USER_DATA_ADDR_LO_GS] = HwShIgnoreRegisters;
 	g_hw_sh_func[Pm4::SPI_SHADER_USER_DATA_ADDR_HI_GS] = HwShIgnoreRegisters;
-	g_hw_sh_func[Pm4::SPI_SHADER_REQ_CTRL_ESGS]        = HwShIgnoreRegisters;
-	g_hw_sh_func[Pm4::SPI_SHADER_PGM_RSRC3_HS]         = HwShIgnoreRegisters;
-	g_hw_sh_func[Pm4::SPI_SHADER_PGM_RSRC4_HS]         = HwShIgnoreRegisters;
+	g_hw_sh_func[Pm4::SPI_GRAPHICS_SHADER_CONTROL_HS]  = HwShIgnoreRegisters;
 	g_hw_sh_func[Pm4::SPI_SHADER_USER_DATA_ADDR_LO_HS] = HwShIgnoreRegisters;
 	g_hw_sh_func[Pm4::SPI_SHADER_USER_DATA_ADDR_HI_HS] = HwShIgnoreRegisters;
-	g_hw_sh_func[Pm4::SPI_SHADER_REQ_CTRL_LSHS]        = HwShIgnoreRegisters;
 
 	return g_hw_sh_func;
 }
@@ -179,6 +158,7 @@ constexpr auto MakeUconfigDispatchTable() {
 	g_hw_uc_func[Pm4::VGT_INDEX_TYPE]            = HwUcSetIndexType;
 	g_hw_uc_func[Pm4::VGT_OBJECT_ID]             = HwUcSetObjectId;
 	g_hw_uc_func[Pm4::TEXTURE_GRADIENT_FACTORS]  = HwUcSetTextureGradientFactors;
+	g_hw_uc_func[Pm4::TEXTURE_GRADIENT_CONTROL]  = HwUcSetTextureGradientFactors;
 	g_hw_uc_func[Pm4::IA_MULTI_VGT_PARAM]        = HwUcSetIaMultiVgtParam;
 	g_hw_uc_func[Pm4::GE_MULTI_PRIM_IB_RESET_EN] = HwUcSetMultiPrimIbReset;
 	g_hw_uc_func[Pm4::TA_CS_BC_BASE_ADDR]        = HwUcSetBorderColorTableAddr;

src/graphics/guest_gpu/command_processor/pm4Dispatch.h

@@ -44,11 +44,9 @@ uint32_t HwCtxSetBorderColorTableAddr(CommandProcessor&, uint32_t, uint32_t, con
 uint32_t HwCtxSetStencilClear(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
 uint32_t HwCtxSetDepthClear(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
 uint32_t HwCtxSetScreenScissor(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
-uint32_t HwCtxSetDepthRenderTarget(CommandProcessor&, uint32_t, uint32_t, const uint32_t*,
-                                   uint32_t);
 uint32_t HwCtxSetStencilInfo(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
-uint32_t HwCtxSetDepthHtileSurface(CommandProcessor&, uint32_t, uint32_t, const uint32_t*,
-                                   uint32_t);
+uint32_t HwCtxSetDepthShadingRateEncoding(CommandProcessor&, uint32_t, uint32_t,
+                                          const uint32_t*, uint32_t);
 uint32_t HwCtxSetHardwareScreenOffset(CommandProcessor&, uint32_t, uint32_t, const uint32_t*,
                                       uint32_t);
 uint32_t HwCtxSetWindowOffset(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
@@ -89,7 +87,6 @@ uint32_t HwCtxSetDrawPayloadControl(CommandProcessor&, uint32_t, uint32_t, const
 uint32_t HwCtxSetPrimitiveIdReset(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
 uint32_t HwCtxSetObjprimIdControl(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
 uint32_t HwCtxSetShaderStages(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
-uint32_t HwCtxSetRenderTarget(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
 uint32_t HwCtxSetColorInfo(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
 uint32_t HwCtxSetBlendControl(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
 uint32_t HwCtxSetViewportScissor(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
@@ -100,10 +97,8 @@ uint32_t HwShSetPsUserSgpr(CommandProcessor&, uint32_t, uint32_t, const uint32_t
 uint32_t HwShSetGsUserSgpr(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
 uint32_t HwShSetHsUserSgpr(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
 uint32_t HwShSetCsUserSgpr(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
-uint32_t HwShSetPsUserAccumSgpr(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
-uint32_t HwShSetGsUserAccumSgpr(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
-uint32_t HwShSetHsUserAccumSgpr(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
-uint32_t HwShSetCsUserAccumSgpr(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
+uint32_t HwShIgnoreUserAccumulator(CommandProcessor&, uint32_t, uint32_t, const uint32_t*,
+                                   uint32_t);
 uint32_t HwShSetCsRegisters(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
 uint32_t HwShIgnoreRegisters(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);
 uint32_t HwUcSetPrimitiveType(CommandProcessor&, uint32_t, uint32_t, const uint32_t*, uint32_t);

src/graphics/guest_gpu/gpu_defs.h

@@ -407,7 +407,19 @@ enum class BufferFormat : uint32_t {
 	k5_5_5_1UNorm       = 134,
 	k1_5_5_5UNorm       = 135,
 	k4_4_4_4UNorm       = 136,
+	kFmask8_S2_F1       = 156,
+	kFmask8_S4_F1       = 157,
+	kFmask8_S8_F1       = 158,
+	kFmask8_S2_F2       = 159,
+	kFmask8_S4_F2       = 160,
 	kFmask8_S4_F4       = 161,
+	kFmask16_S16_F1     = 162,
+	kFmask16_S8_F2      = 163,
+	kFmask32_S16_F2     = 164,
+	kFmask32_S8_F4      = 165,
+	kFmask32_S8_F8      = 166,
+	kFmask64_S16_F4     = 167,
+	kFmask64_S16_F8     = 168,
 	kBc1UNorm           = 169,
 	kBc1Srgb            = 170,
 	kBc2UNorm           = 171,

src/graphics/guest_gpu/gpu_format.cpp

@@ -68,7 +68,19 @@ constexpr FormatInfo kFormatInfo[] = {
 	{BufferFormat::k5_5_5_1UNorm, 2, 0, 2, true, false},
 	{BufferFormat::k1_5_5_5UNorm, 0, 0, 2, false, false},
 	{BufferFormat::k4_4_4_4UNorm, 2, 0, 2, true, false},
+	{BufferFormat::kFmask8_S2_F1, 1, 0, 1, true, false},
+	{BufferFormat::kFmask8_S4_F1, 1, 0, 1, true, false},
+	{BufferFormat::kFmask8_S8_F1, 1, 0, 1, true, false},
+	{BufferFormat::kFmask8_S2_F2, 1, 0, 1, true, false},
+	{BufferFormat::kFmask8_S4_F2, 1, 0, 1, true, false},
 	{BufferFormat::kFmask8_S4_F4, 1, 0, 1, true, false},
+	{BufferFormat::kFmask16_S16_F1, 2, 0, 2, true, false},
+	{BufferFormat::kFmask16_S8_F2, 2, 0, 2, true, false},
+	{BufferFormat::kFmask32_S16_F2, 4, 0, 4, true, false},
+	{BufferFormat::kFmask32_S8_F4, 4, 0, 4, true, false},
+	{BufferFormat::kFmask32_S8_F8, 4, 0, 4, true, false},
+	{BufferFormat::kFmask64_S16_F4, 8, 0, 8, true, false},
+	{BufferFormat::kFmask64_S16_F8, 8, 0, 8, true, false},
 	{BufferFormat::kBc1UNorm, 0, 8, 0, true, false},
 	{BufferFormat::kBc1Srgb, 0, 8, 0, true, false},
 	{BufferFormat::kBc2UNorm, 0, 16, 0, true, false},
@@ -203,7 +215,7 @@ uint32_t RenderTargetBytesPerElement(BufferFormat format) {
 }
 
 bool IsFmaskTextureFormat(BufferFormat format) {
-	return format == BufferFormat::kFmask8_S4_F4;
+	return format >= BufferFormat::kFmask8_S2_F1 && format <= BufferFormat::kFmask64_S16_F8;
 }
 
 bool IsSampledTextureFormat(BufferFormat format) {

src/graphics/guest_gpu/graphicsRun.cpp

@@ -27,6 +27,7 @@
 #include <cstdio>
 #include <deque>
 #include <memory>
+#include <mutex>
 #include <semaphore>
 #include <thread>
 #include <vector>
@@ -65,17 +66,10 @@ static bool GraphicsRunDebugDumpEnabled() {
 	       Config::GetPrintfDirection() != Config::OutputDirection::Silent;
 }
 
-GuestGpu::OwnedCmdBuffer::OwnedCmdBuffer(const uint32_t* data, uint32_t count) {
-	EXIT_IF(data == nullptr && count != 0);
-	if (count != 0) {
-		m_words.assign(data, data + count);
-	}
-}
-
 GuestGpu::GuestGpu(RenderContext& renderer): m_renderer(renderer) {
 	EXIT_NOT_IMPLEMENTED(!Common::Thread::IsMainThread());
 	GraphicsInitJmpTables();
-	m_gfx_cp = std::make_unique<CommandProcessor>(renderer);
+	m_gfx_cp = std::make_unique<CommandProcessor>(renderer, 0);
 	m_thread = std::jthread(ThreadRun, this);
 }
 
@@ -147,39 +141,33 @@ void GuestGpu::SendCommandSync(Common::UniqueFunction<void>&& command) {
 	done.acquire();
 }
 
-void GuestGpu::Submit(uint32_t* cmd_draw_buffer, uint32_t num_draw_dw, uint32_t* cmd_const_buffer,
-                      uint32_t num_const_dw, bool trigger_agc_interrupt_on_done) {
-	if (num_draw_dw == 0) {
+void GuestGpu::Submit(std::span<const uint32_t> draw_commands,
+                      std::span<const uint32_t> constant_commands) {
+	if (draw_commands.empty()) {
 		return;
 	}
-	EXIT_IF(cmd_draw_buffer == nullptr);
 	GpuMutexLock lock(m_submission_mutex);
 	Submission   submission;
-	submission.type                          = SubmissionType::Graphics;
-	submission.queue_id                      = 0;
-	submission.commands                      = OwnedCmdBuffer(cmd_draw_buffer, num_draw_dw);
-	submission.constant_commands             = OwnedCmdBuffer(cmd_const_buffer, num_const_dw);
-	submission.trigger_agc_interrupt_on_done = trigger_agc_interrupt_on_done;
-	submission.reset_processor               = m_graphics_done;
-	m_graphics_done                          = false;
+	submission.type              = SubmissionType::Graphics;
+	submission.queue_id          = 0;
+	submission.commands          = draw_commands;
+	submission.constant_commands = constant_commands;
+	submission.reset_processor   = m_graphics_done;
+	m_graphics_done              = false;
 	Enqueue(std::move(submission));
 }
 
-void GuestGpu::SubmitCompute(uint32_t queue, uint32_t* cmd_buffer, uint32_t num_dw,
-                             bool trigger_agc_interrupt_on_done) {
-	EXIT_IF(cmd_buffer == nullptr || num_dw == 0);
+void GuestGpu::SubmitCompute(uint32_t queue, std::span<const uint32_t> commands) {
+	EXIT_IF(commands.empty());
 	GpuMutexLock lock(m_submission_mutex);
 
-	constexpr uint32_t compute_queue_base = 0x20u;
-	EXIT_NOT_IMPLEMENTED(queue < compute_queue_base ||
-	                     queue >= compute_queue_base + ComputeQueueCount);
+	EXIT_NOT_IMPLEMENTED(queue < ComputeQueueBase || queue >= ComputeQueueBase + ComputeQueueCount);
 
-	const auto compute_queue = queue - compute_queue_base;
+	const auto compute_queue = queue - ComputeQueueBase;
 	Submission submission;
-	submission.type                          = SubmissionType::Compute;
-	submission.queue_id                      = 1 + compute_queue;
-	submission.commands                      = OwnedCmdBuffer(cmd_buffer, num_dw);
-	submission.trigger_agc_interrupt_on_done = trigger_agc_interrupt_on_done;
+	submission.type     = SubmissionType::Compute;
+	submission.queue_id = 1 + compute_queue;
+	submission.commands = commands;
 	Enqueue(std::move(submission));
 }
 
@@ -214,7 +202,7 @@ CommandProcessor& GuestGpu::GetProcessor(uint32_t queue_id) {
 	}
 	auto& processor = m_compute_cp[queue_id - 1];
 	if (processor == nullptr) {
-		processor = std::make_unique<CommandProcessor>(m_renderer);
+		processor = std::make_unique<CommandProcessor>(m_renderer, ComputeQueueBase + queue_id - 1);
 	}
 	return *processor;
 }
@@ -267,8 +255,7 @@ void CommandProcessor::BufferFlush() {
 }
 
 void CommandProcessor::BufferFlushAndWait() {
-	auto& submitted = GetScheduler().FlushAndGetSubmitted();
-	submitted.WaitForFenceOnly();
+	GetScheduler().FlushAndWait();
 }
 
 void CommandProcessor::BufferWait() {
@@ -582,21 +569,20 @@ bool GuestGpu::Process(Submission& submission) {
 	switch (submission.type) {
 		case SubmissionType::Graphics: {
 			bool progressed = false;
-			submission.constant_complete |= submission.constant_commands.Empty();
+			submission.constant_complete |= submission.constant_commands.empty();
 			for (;;) {
 				bool round_progress = false;
 				if (!submission.constant_complete) {
 					submission.constant_complete =
-					    cp.Process(
-					        submission.constant_execution, submission.constant_commands.Data(),
-					        submission.constant_commands.Size()) == Pm4ProcessResult::Complete;
+					    cp.Process(submission.constant_execution, submission.constant_commands) ==
+					    Pm4ProcessResult::Complete;
 					round_progress |= submission.constant_execution.MadeProgress();
 				}
 				cp.SetCeComplete(submission.constant_complete);
 				if (!submission.command_complete) {
 					submission.command_complete =
-					    cp.Process(submission.command_execution, submission.commands.Data(),
-					               submission.commands.Size()) == Pm4ProcessResult::Complete;
+					    cp.Process(submission.command_execution, submission.commands) ==
+					    Pm4ProcessResult::Complete;
 					round_progress |= submission.command_execution.MadeProgress();
 				}
 				progressed |= round_progress;
@@ -613,14 +599,11 @@ bool GuestGpu::Process(Submission& submission) {
 			} else if (complete) {
 				m_renderer.GetGpuResources().RunGarbageCollector();
 			}
-			if (complete && submission.trigger_agc_interrupt_on_done) {
-				m_renderer.TriggerEopEvent(0);
-			}
 			break;
 		}
 		case SubmissionType::Compute: {
-			const auto      num_dw                  = submission.commands.Size();
-			auto*           buffer                  = submission.commands.Data();
+			const auto      num_dw = static_cast<uint32_t>(submission.commands.size());
+			const auto*     buffer = submission.commands.data();
 			static uint32_t compute_batch_log_count = 0;
 			if (first_slice && num_dw <= 128 && compute_batch_log_count++ < 32) {
 				LOGF("compute direct batch: data=0x%016" PRIx64 ", num_dw=%" PRIu32 "\n",
@@ -632,7 +615,7 @@ bool GuestGpu::Process(Submission& submission) {
 			if (first_slice) {
 				GraphicsDbgDumpDcb("cc", num_dw, buffer);
 			}
-			complete = cp.Process(submission.command_execution, buffer, num_dw) ==
+			complete = cp.Process(submission.command_execution, submission.commands) ==
 			           Pm4ProcessResult::Complete;
 			if (submission.command_execution.MadeProgress()) {
 				if (complete) {
@@ -642,9 +625,6 @@ bool GuestGpu::Process(Submission& submission) {
 			} else if (complete) {
 				m_renderer.GetGpuResources().RunGarbageCollector();
 			}
-			if (complete && submission.trigger_agc_interrupt_on_done) {
-				Sync::TriggerAgcUserInterrupt();
-			}
 			break;
 		}
 		case SubmissionType::FlipPreparation:
@@ -656,12 +636,13 @@ bool GuestGpu::Process(Submission& submission) {
 	return complete;
 }
 
-Pm4ProcessResult CommandProcessor::Process(Pm4Execution& execution, uint32_t* buffer,
-                                           uint32_t size_dw) {
+Pm4ProcessResult CommandProcessor::Process(Pm4Execution&             execution,
+                                           std::span<const uint32_t> commands) {
 	KYTY_PROFILER_BLOCK("CommandProcessor::Process");
 	EXIT_IF(g_current_execution != nullptr);
-	if (execution.m_buffer_stack.empty() && size_dw != 0) {
-		execution.m_buffer_stack.push_back({buffer, size_dw, size_dw, 0});
+	EXIT_IF(commands.size() > UINT32_MAX);
+	if (execution.m_buffer_stack.empty() && !commands.empty()) {
+		execution.m_buffer_stack.push_back({commands});
 	}
 	execution.m_suspended     = false;
 	execution.m_made_progress = false;
@@ -686,14 +667,14 @@ Pm4ProcessResult CommandProcessor::Process(Pm4Execution& execution, uint32_t* bu
 	                                        : Pm4ProcessResult::Blocked;
 }
 
-void CommandProcessor::ProcessIndirectBuffer(uint32_t* buffer, uint32_t size_dw) {
+void CommandProcessor::ProcessIndirectBuffer(std::span<const uint32_t> commands) {
 	EXIT_IF(g_current_execution == nullptr);
-	if (size_dw == 0) {
+	if (commands.empty()) {
 		return;
 	}
 	auto&      execution  = *g_current_execution;
 	const auto stop_depth = execution.m_buffer_stack.size();
-	execution.m_buffer_stack.push_back({buffer, size_dw, size_dw, 0});
+	execution.m_buffer_stack.push_back({commands});
 	ProcessPm4(execution, stop_depth);
 }
 
@@ -709,29 +690,28 @@ void CommandProcessor::ProcessPm4(Pm4Execution& execution, size_t stop_depth) {
 		}
 		const auto buffer_index = execution.m_buffer_stack.size() - 1;
 		auto&      cursor       = execution.m_buffer_stack[buffer_index];
+		EXIT_IF(cursor.offset_dw > cursor.commands.size());
 		if (cursor.deferred_advance_dw != 0) {
-			EXIT_IF(cursor.deferred_advance_dw > cursor.remaining_dw);
-			cursor.next_packet += cursor.deferred_advance_dw;
-			cursor.remaining_dw -= cursor.deferred_advance_dw;
+			EXIT_IF(cursor.deferred_advance_dw > cursor.commands.size() - cursor.offset_dw);
+			cursor.offset_dw += cursor.deferred_advance_dw;
 			cursor.deferred_advance_dw = 0;
 			execution.m_made_progress  = true;
 			continue;
 		}
-		if (cursor.remaining_dw == 0) {
+		if (cursor.offset_dw == cursor.commands.size()) {
 			execution.m_buffer_stack.pop_back();
 			continue;
 		}
 
-		auto* const packet        = cursor.next_packet;
-		const auto  remaining_dw  = cursor.remaining_dw;
-		const auto  total_dw      = cursor.total_dw;
-		const auto  packet_header = packet[0];
-		const auto  opcode        = (packet_header >> 8u) & 0xffu;
+		const auto* const packet        = cursor.commands.data() + cursor.offset_dw;
+		const auto        total_dw      = static_cast<uint32_t>(cursor.commands.size());
+		const auto        remaining_dw  = total_dw - cursor.offset_dw;
+		const auto        packet_header = packet[0];
+		const auto        opcode        = (packet_header >> 8u) & 0xffu;
 		EXIT_NOT_IMPLEMENTED(remaining_dw > total_dw);
 
 		if (packet_header == 0x80000000u) {
-			cursor.next_packet++;
-			cursor.remaining_dw--;
+			cursor.offset_dw++;
 			execution.m_made_progress = true;
 			continue;
 		}
@@ -766,8 +746,7 @@ void CommandProcessor::ProcessPm4(Pm4Execution& execution, size_t stop_depth) {
 					     dst, val, packet[1], packet[2]);
 				}
 			}
-			cursor.next_packet += packet_dw;
-			cursor.remaining_dw -= packet_dw;
+			cursor.offset_dw += packet_dw;
 			execution.m_made_progress = true;
 			continue;
 		}
@@ -800,8 +779,7 @@ void CommandProcessor::ProcessPm4(Pm4Execution& execution, size_t stop_depth) {
 			return;
 		}
 		EXIT_IF(execution.m_buffer_stack.size() != buffer_index + 1);
-		execution.m_buffer_stack[buffer_index].next_packet += packet_dw;
-		execution.m_buffer_stack[buffer_index].remaining_dw -= packet_dw;
+		execution.m_buffer_stack[buffer_index].offset_dw += packet_dw;
 		execution.m_made_progress = true;
 	}
 }
@@ -1244,7 +1222,14 @@ void CommandProcessor::WriteAtEndOfPipe(uint32_t cache_policy, uint32_t event_wr
 	switch (interrupt_selector) {
 		case 0x00:
 		case 0x03: with_interrupt = false; break;
-		case 0x01: Sync::TriggerEopEventAtEndOfPipe(CurrentBuffer(), interrupt_context_id); return;
+		case 0x01:
+			if (!IsAsyncComputeQueue()) {
+				Sync::TriggerEopEventAtEndOfPipe(CurrentBuffer(), m_interrupt_event_id,
+				                                 interrupt_context_id);
+				return;
+			}
+			with_interrupt = true;
+			break;
 		case 0x02: with_interrupt = true; break;
 		default: EXIT("unknown interrupt_selector\n");
 	}
@@ -1257,10 +1242,11 @@ void CommandProcessor::WriteAtEndOfPipe(uint32_t cache_policy, uint32_t event_wr
 		if (with_interrupt) {
 			if (with_writeback) {
 				Sync::WriteAtEndOfPipeWithInterruptWriteBack32(m_submit_id, CurrentBuffer(), dst,
-				                                               data, interrupt_context_id);
+				                                               data, m_interrupt_event_id,
+				                                               interrupt_context_id);
 			} else {
 				Sync::WriteAtEndOfPipeWithInterrupt32(m_submit_id, CurrentBuffer(), dst, data,
-				                                      interrupt_context_id);
+				                                      m_interrupt_event_id, interrupt_context_id);
 			}
 		} else if (with_writeback) {
 			Sync::WriteAtEndOfPipeWithWriteBack32(m_submit_id, CurrentBuffer(), dst, data);
@@ -1275,12 +1261,12 @@ void CommandProcessor::WriteAtEndOfPipe(uint32_t cache_policy, uint32_t event_wr
 				if (eop_event_type == 0x2f && cache_action == 0x00 && event_index == 0x06) {
 					auto* dst = static_cast<uint32_t*>(dst_gpu_addr);
 					SynchronizeGpu();
-					Sync::ReadGds(m_renderer.GetBufferCache().GetGdsBuffer(), dst, value & 0xffffu,
+					Sync::ReadGds(*m_renderer.GetBufferCache().GetGdsBuffer(), dst, value & 0xffffu,
 					              value >> 16u);
 					Sync::WriteAtEndOfPipeGds32(m_submit_id, CurrentBuffer(), dst, value & 0xffffu,
 					                            value >> 16u);
 					if (with_interrupt) {
-						m_renderer.TriggerEopEvent(interrupt_context_id);
+						m_renderer.TriggerInterrupt(m_interrupt_event_id, interrupt_context_id);
 					}
 					return;
 				}
@@ -1306,10 +1292,12 @@ void CommandProcessor::WriteAtEndOfPipe(uint32_t cache_policy, uint32_t event_wr
 					if (with_interrupt) {
 						if (with_writeback) {
 							Sync::WriteAtEndOfPipeWithInterruptWriteBack64(
-							    m_submit_id, CurrentBuffer(), dst, value, interrupt_context_id);
+							    m_submit_id, CurrentBuffer(), dst, value, m_interrupt_event_id,
+							    interrupt_context_id);
 						} else {
 							Sync::WriteAtEndOfPipeWithInterrupt64(m_submit_id, CurrentBuffer(), dst,
-							                                      value, interrupt_context_id);
+							                                      value, m_interrupt_event_id,
+							                                      interrupt_context_id);
 						}
 					} else if (with_writeback) {
 						Sync::WriteAtEndOfPipeWithWriteBack64(m_submit_id, CurrentBuffer(), dst,
@@ -1388,26 +1376,35 @@ void CommandProcessor::WriteAtEndOfPipe(uint32_t cache_policy, uint32_t event_wr
 		case 0x04:
 			if constexpr (sizeof(T) == sizeof(uint64_t)) {
 				const auto clock = Sync::ReadReferenceClock();
-				std::memcpy(dst_gpu_addr, &clock, sizeof(clock));
+				auto*      dst   = static_cast<uint64_t*>(dst_gpu_addr);
+				std::memcpy(dst, &clock, sizeof(clock));
 				switch (cache_action) {
 					case 0x00:
-						if (((eop_event_type == 0x04 && event_index == 0x05) ||
-						     (eop_event_type == 0x28 && event_index == 0x00)) &&
-						    !with_interrupt) {
-							Sync::WriteAtEndOfPipeClockCounter(m_submit_id, CurrentBuffer(),
-							                                   static_cast<uint64_t*>(dst_gpu_addr),
-							                                   clock);
+						if ((eop_event_type == 0x04 && event_index == 0x05) ||
+						    (eop_event_type == 0x28 && event_index == 0x00)) {
+							if (with_interrupt) {
+								Sync::WriteAtEndOfPipeWithInterrupt64(
+								    m_submit_id, CurrentBuffer(), dst, clock, m_interrupt_event_id,
+								    interrupt_context_id);
+							} else {
+								Sync::WriteAtEndOfPipeClockCounter(m_submit_id, CurrentBuffer(),
+								                                   dst, clock);
+							}
 							return;
 						}
 						break;
 					case 0x38:
-						if (((eop_event_type == 0x04 &&
-						      (event_index == 0x00 || event_index == 0x05)) ||
-						     (eop_event_type == 0x28 && event_index == 0x00)) &&
-						    !with_interrupt) {
-							Sync::WriteAtEndOfPipeClockCounterWithWriteBack(
-							    m_submit_id, CurrentBuffer(), static_cast<uint64_t*>(dst_gpu_addr),
-							    clock);
+						if ((eop_event_type == 0x04 &&
+						     (event_index == 0x00 || event_index == 0x05)) ||
+						    (eop_event_type == 0x28 && event_index == 0x00)) {
+							if (with_interrupt) {
+								Sync::WriteAtEndOfPipeWithInterruptWriteBack64(
+								    m_submit_id, CurrentBuffer(), dst, clock, m_interrupt_event_id,
+								    interrupt_context_id);
+							} else {
+								Sync::WriteAtEndOfPipeClockCounterWithWriteBack(
+								    m_submit_id, CurrentBuffer(), dst, clock);
+							}
 							return;
 						}
 						break;
@@ -1464,15 +1461,17 @@ void CommandProcessor::EmitGlobalBarrier() {
 void CommandProcessor::TriggerEopEventAtEndOfPipe(uint32_t interrupt_context_id) {
 	CheckBuffer();
 
-	Sync::TriggerEopEventAtEndOfPipe(CurrentBuffer(), interrupt_context_id);
+	Sync::TriggerEopEventAtEndOfPipe(CurrentBuffer(), m_interrupt_event_id, interrupt_context_id);
 }
 
-void CommandProcessor::TriggerEvent(uint32_t event_type, uint32_t event_index) {
+void CommandProcessor::TriggerEvent(uint32_t event_type, uint32_t event_index,
+                                    uint64_t event_address) {
 	if (GraphicsRunDebugDumpEnabled()) {
 		LOGF("CommandProcessor::TriggerEvent()\n"
 		     "\t event_type  = 0x%08" PRIx32 "\n"
-		     "\t event_index = 0x%08" PRIx32 "\n",
-		     event_type, event_index);
+		     "\t event_index = 0x%08" PRIx32 "\n"
+		     "\t address     = 0x%016" PRIx64 "\n",
+		     event_type, event_index, event_address);
 	}
 
 	const auto valid_cache_event_index = event_index == 0x00000000 || event_index == 0x00000007;
@@ -1509,12 +1508,36 @@ void CommandProcessor::TriggerEvent(uint32_t event_type, uint32_t event_index) {
 		case 0x0000001a:
 		case 0x0000001b:
 		case 0x00000038:
-		case 0x00000039:
 		case 0x0000003a:
 			LOGF("\t temporary: ignoring unsupported event_write type 0x%08" PRIx32
 			     ", index 0x%08" PRIx32 "\n",
 			     event_type, event_index);
 			break;
+		case 0x00000039: {
+			if (event_index != 0x00000001 || event_address == 0 || (event_address & 0x7u) != 0) {
+				EXIT("invalid occlusion-counter dump: index=0x%08" PRIx32 ", address=0x%016" PRIx64
+				     "\n",
+				     event_index, event_address);
+			}
+			static std::once_flag warning_once;
+			std::call_once(warning_once, [] {
+				std::printf("Warning: game uses occlusion queries, which are currently treated as "
+				            "always visible; GPU usage may be higher and FPS may be lower.\n");
+			});
+
+			// Until host occlusion queries are implemented, publish an always-visible result. The
+			// PS5 layout contains one interleaved begin/end pair per DB, and bit 63 marks a result
+			// ready.
+			constexpr uint64_t ready_bit    = 1ull << 63u;
+			constexpr uint64_t counter_mask = ready_bit - 1u;
+			auto*              results      = reinterpret_cast<volatile uint64_t*>(event_address);
+			const auto         value        = ready_bit | m_synthetic_occlusion_counter;
+			for (uint32_t db = 0; db < 16u; db++) {
+				results[db * 2u] = value;
+			}
+			m_synthetic_occlusion_counter = (m_synthetic_occlusion_counter + 1u) & counter_mask;
+			break;
+		}
 		default:
 			EXIT("unknown event type: 0x%08" PRIx32 ", 0x%08" PRIx32 "\n", event_type, event_index);
 	}
@@ -1577,7 +1600,7 @@ void CommandProcessor::FlipWithInterrupt(uint32_t eop_event_type, uint32_t cache
 	                                         m_flip.flip_arg);
 	Sync::WriteAtEndOfPipeWithInterruptWriteBackFlip32(
 	    m_submit_id, command, static_cast<uint32_t*>(dst_gpu_addr), value, m_flip.handle,
-	    m_flip.index, m_flip.flip_mode, m_flip.flip_arg, request);
+	    m_flip.index, m_flip.flip_mode, m_flip.flip_arg, request, m_interrupt_event_id);
 	GetScheduler().Flush();
 }
 
@@ -1598,7 +1621,7 @@ void CommandProcessor::PrepareCpuFlip(uint64_t request_id) {
 }
 
 void CommandProcessor::SynchronizeGpu() {
-	GetScheduler().FinishCurrent();
+	GetScheduler().Finish();
 }
 
 bool GuestGpu::IsGpuThread() noexcept {

src/graphics/guest_gpu/graphicsRun.h

@@ -13,8 +13,8 @@
 #include <deque>
 #include <memory>
 #include <mutex>
+#include <span>
 #include <thread>
-#include <vector>
 
 namespace Libs::Graphics {
 
@@ -31,12 +31,12 @@ public:
 	void               SendCommand(Common::UniqueFunction<void>&& command);
 	void               SendCommandSync(Common::UniqueFunction<void>&& command);
 
-	void Submit(uint32_t* draw_commands, uint32_t draw_size_dw, uint32_t* constant_commands,
-	            uint32_t constant_size_dw, bool trigger_agc_interrupt_on_done = false);
-	void SubmitCompute(uint32_t queue, uint32_t* commands, uint32_t size_dw,
-	                   bool trigger_agc_interrupt_on_done = false);
-	void SubmitFlipPreparation(uint64_t request_id);
-	void Done();
+	// Submitted command memory is borrowed and must remain valid until GPU execution completes.
+	void              Submit(std::span<const uint32_t> draw_commands,
+	                         std::span<const uint32_t> constant_commands);
+	void              SubmitCompute(uint32_t queue, std::span<const uint32_t> commands);
+	void              SubmitFlipPreparation(uint64_t request_id);
+	void              Done();
 	[[nodiscard]] int GetFrameNum() const;
 
 	[[nodiscard]] static bool IsGpuThread() noexcept;
@@ -45,39 +45,24 @@ private:
 	static constexpr uint32_t ComputePipeCount     = 7;
 	static constexpr uint32_t QueuesPerComputePipe = 8;
 	static constexpr uint32_t ComputeQueueCount    = ComputePipeCount * QueuesPerComputePipe;
+	static constexpr uint32_t ComputeQueueBase     = 0x20;
 	static constexpr uint32_t QueueCount           = 1 + ComputeQueueCount;
 
-	class OwnedCmdBuffer {
-	public:
-		OwnedCmdBuffer() = default;
-		explicit OwnedCmdBuffer(const uint32_t* data, uint32_t count);
-
-		[[nodiscard]] bool     Empty() const noexcept { return m_words.empty(); }
-		[[nodiscard]] uint32_t Size() const noexcept {
-			return static_cast<uint32_t>(m_words.size());
-		}
-		[[nodiscard]] uint32_t* Data() noexcept { return m_words.data(); }
-
-	private:
-		std::vector<uint32_t> m_words;
-	};
-
 	enum class SubmissionType { Graphics, Compute, FlipPreparation };
 
 	struct Submission {
-		SubmissionType type     = SubmissionType::Graphics;
-		uint32_t       queue_id = 0;
-		OwnedCmdBuffer commands;
-		OwnedCmdBuffer constant_commands;
-		Pm4Execution   command_execution;
-		Pm4Execution   constant_execution;
-		bool           trigger_agc_interrupt_on_done = false;
-		bool           reset_processor               = false;
-		bool           started                       = false;
-		bool           command_complete              = false;
-		bool           constant_complete             = false;
-		bool           blocked                       = false;
-		uint64_t       flip_request_id               = 0;
+		SubmissionType            type     = SubmissionType::Graphics;
+		uint32_t                  queue_id = 0;
+		std::span<const uint32_t> commands;
+		std::span<const uint32_t> constant_commands;
+		Pm4Execution              command_execution;
+		Pm4Execution              constant_execution;
+		bool                      reset_processor   = false;
+		bool                      started           = false;
+		bool                      command_complete  = false;
+		bool                      constant_complete = false;
+		bool                      blocked           = false;
+		uint64_t                  flip_request_id   = 0;
 	};
 
 	void              Enqueue(Submission submission);

src/graphics/guest_gpu/hardwareContext.h

@@ -11,15 +11,6 @@ struct ColorBase {
 	uint64_t addr = 0;
 };
 
-struct ColorPitch {
-	uint32_t pitch_div8_minus1       = 0;
-	uint32_t fmask_pitch_div8_minus1 = 0;
-};
-
-struct ColorSlice {
-	uint32_t slice_div64_minus1 = 0;
-};
-
 struct ColorView {
 	uint32_t base_array_slice_index = 0;
 	uint32_t last_array_slice_index = 0;
@@ -36,20 +27,15 @@ struct ColorInfo {
 	bool                    blend_clamp                    = false;
 	bool                    blend_bypass                   = false;
 	bool                    round_mode                     = false;
-	uint32_t                cmask_is_linear                = 0;
-	uint32_t                cmask_addr_type                = 0;
-	bool                    alt_tile_mode                  = false;
 	Prospero::ChannelLayout format                         = Prospero::ChannelLayout::kInvalid;
 	Prospero::ChannelType   channel_type                   = Prospero::ChannelType::kUNorm;
 	Prospero::ChannelOrder  channel_order                  = Prospero::ChannelOrder::kStandard;
 };
 
 struct ColorAttrib {
-	bool               force_dest_alpha_to_one = false;
-	Prospero::TileMode tile_mode               = Prospero::TileMode::kLinear;
-	Prospero::TileMode fmask_tile_mode         = Prospero::TileMode::kLinear;
-	uint32_t           num_samples             = 0;
-	uint32_t           num_fragments           = 0;
+	bool     force_dest_alpha_to_one = false;
+	uint32_t num_samples             = 0;
+	uint32_t num_fragments           = 0;
 };
 
 struct ColorAttrib2 {
@@ -59,41 +45,33 @@ struct ColorAttrib2 {
 };
 
 struct ColorAttrib3 {
-	uint32_t           depth              = 0;
-	Prospero::TileMode tile_mode          = Prospero::TileMode::kLinear;
-	uint32_t           dimension          = 0;
-	bool               cmask_pipe_aligned = false;
-	bool               dcc_pipe_aligned   = false;
+	uint32_t           depth                        = 0;
+	Prospero::TileMode tile_mode                    = Prospero::TileMode::kRenderTarget;
+	uint32_t           dimension                    = 0;
+	bool               metadata_pipe_aligned        = false;
+	bool               write_vrs_rate_hint_to_cmask = false;
 };
 
 struct ColorDccControl {
-	uint32_t max_uncompressed_block_size    = 0;
-	uint32_t max_compressed_block_size      = 0;
-	uint32_t min_compressed_block_size      = 0;
-	uint32_t color_transform                = 0;
-	bool     dcc_clear_key_enable           = false;
-	bool     overwrite_combiner_disable     = false;
-	bool     independent_64b_blocks         = false;
-	bool     independent_128b_blocks        = false;
-	bool     data_write_on_dcc_clear_to_reg = false;
+	enum class IndependentBlockSize : uint8_t { Disabled, Bytes64, Bytes128 };
+
+	uint32_t             max_uncompressed_block_size    = 2;
+	uint32_t             max_compressed_block_size      = 2;
+	uint32_t             color_transform                = 0;
+	bool                 dcc_clear_key_enable           = false;
+	bool                 overwrite_combiner_disable     = false;
+	IndependentBlockSize independent_block_size         = IndependentBlockSize::Disabled;
+	bool                 data_write_on_dcc_clear_to_reg = false;
 };
 
 struct ColorCmask {
 	uint64_t addr = 0;
 };
 
-struct ColorCmaskSlice {
-	uint32_t slice_minus1 = 0;
-};
-
 struct ColorFmask {
 	uint64_t addr = 0;
 };
 
-struct ColorFmaskSlice {
-	uint32_t slice_minus1 = 0;
-};
-
 struct ColorClearWord0 {
 	uint32_t word0 = 0;
 };
@@ -106,15 +84,8 @@ struct ColorDccAddr {
 	uint64_t addr = 0;
 };
 
-struct ColorSize {
-	uint32_t width  = 0;
-	uint32_t height = 0;
-};
-
 struct RenderTarget {
 	ColorBase       base;
-	ColorPitch      pitch;
-	ColorSlice      slice;
 	ColorView       view;
 	ColorInfo       info;
 	ColorAttrib     attrib;
@@ -122,13 +93,10 @@ struct RenderTarget {
 	ColorAttrib3    attrib3;
 	ColorDccControl dcc;
 	ColorCmask      cmask;
-	ColorCmaskSlice cmask_slice;
 	ColorFmask      fmask;
-	ColorFmaskSlice fmask_slice;
 	ColorClearWord0 clear_word0;
 	ColorClearWord1 clear_word1;
 	ColorDccAddr    dcc_addr;
-	ColorSize       size;
 };
 
 struct DepthZInfo {
@@ -136,7 +104,7 @@ struct DepthZInfo {
 	uint32_t                                    num_samples = 0;
 	Prospero::TextureCompatiblePlaneCompression texture_compatibility =
 	    Prospero::TextureCompatiblePlaneCompression::kDisable;
-	Prospero::ZCompareBase z_compare_base     = Prospero::ZCompareBase::kZMin;
+	Prospero::ZCompareBase z_compare_base     = Prospero::ZCompareBase::kZMax;
 	bool                   htile_acceleration = false;
 	bool                   expclear_enabled   = false;
 	bool                   partially_resident = false;
@@ -171,7 +139,7 @@ struct DepthStencilInfo {
 	Prospero::TextureCompatibleStencil texture_compatibility =
 	    Prospero::TextureCompatibleStencil::kDisable;
 	bool expclear_enabled       = false;
-	bool htile_stencil_disabled = false;
+	bool htile_stencil_disabled = true;
 	bool partially_resident     = false;
 
 	[[nodiscard]] static DepthStencilInfo Decode(uint32_t value) {
@@ -194,16 +162,6 @@ struct DepthStencilInfo {
 	}
 };
 
-struct DepthRenderTargetDepthInfo {
-	uint32_t addr5_swizzle_mask = 0;
-	uint32_t array_mode         = 0;
-	uint32_t pipe_config        = 0;
-	uint32_t bank_width         = 0;
-	uint32_t bank_height        = 0;
-	uint32_t macro_tile_aspect  = 0;
-	uint32_t num_banks          = 0;
-};
-
 struct DepthDepthView {
 	uint32_t slice_start           = 0;
 	uint32_t slice_max             = 0;
@@ -218,36 +176,18 @@ struct DepthDepthSizeXY {
 	bool     valid = false;
 };
 
-struct DepthRenderTargetHTileSurface {
-	uint32_t linear                  = 0;
-	uint32_t full_cache              = 0;
-	uint32_t htile_uses_preload_win  = 0;
-	uint32_t preload                 = 0;
-	uint32_t prefetch_width          = 0;
-	uint32_t prefetch_height         = 0;
-	uint32_t dst_outside_zero_to_one = 0;
-};
-
 struct DepthRenderTarget {
-	DepthZInfo                    z_info;
-	DepthStencilInfo              stencil_info;
-	DepthDepthView                depth_view;
-	DepthDepthSizeXY              size;
-	DepthRenderTargetDepthInfo    depth_info;
-	DepthRenderTargetHTileSurface htile_surface;
+	DepthZInfo       z_info;
+	DepthStencilInfo stencil_info;
+	DepthDepthView   depth_view;
+	DepthDepthSizeXY size;
 
 	uint64_t z_read_base_addr        = 0;
 	uint64_t stencil_read_base_addr  = 0;
 	uint64_t z_write_base_addr       = 0;
 	uint64_t stencil_write_base_addr = 0;
-	uint32_t pitch_div8_minus1       = 0;
-	uint32_t height_div8_minus1      = 0;
-	bool     pitch_height_valid      = false;
-	uint32_t slice_div64_minus1      = 0;
 	uint64_t htile_data_base_addr    = 0;
-	uint32_t width                   = 0;
-	uint32_t height                  = 0;
-	bool     width_height_valid      = false;
+	uint8_t  shading_rate_encoding   = 0;
 };
 
 struct RenderControl {
@@ -256,6 +196,8 @@ struct RenderControl {
 	bool    resummarize_enable       = false;
 	bool    stencil_compress_disable = false;
 	bool    depth_compress_disable   = false;
+	bool    copy_depth_to_color      = false;
+	bool    copy_stencil_to_color    = false;
 	bool    copy_centroid            = false;
 	uint8_t copy_sample              = 0;
 };
@@ -299,16 +241,14 @@ struct ClipControl {
 };
 
 struct DepthControl {
-	bool    stencil_enable                     = false;
-	bool    z_enable                           = false;
-	bool    z_write_enable                     = false;
-	bool    depth_bounds_enable                = false;
-	uint8_t zfunc                              = 0;
-	bool    backface_enable                    = false;
-	uint8_t stencilfunc                        = 0;
-	uint8_t stencilfunc_bf                     = 0;
-	bool    color_writes_on_depth_fail_enable  = false;
-	bool    color_writes_on_depth_pass_disable = false;
+	bool    stencil_enable      = false;
+	bool    z_enable            = false;
+	bool    z_write_enable      = false;
+	bool    depth_bounds_enable = false;
+	uint8_t zfunc               = 0;
+	bool    backface_enable     = false;
+	uint8_t stencilfunc         = 0;
+	uint8_t stencilfunc_bf      = 0;
 };
 
 struct StencilControl {
@@ -468,27 +408,6 @@ struct ScreenViewport {
 	bool     clip_rect_window_offset_enable[4]    = {};
 };
 
-struct VsShaderResource1 {
-	uint8_t vgprs                    = 0;
-	uint8_t sgprs                    = 0;
-	uint8_t priority                 = 0;
-	uint8_t float_mode               = 0;
-	bool    dx10_clamp               = false;
-	bool    ieee_mode                = false;
-	uint8_t vgpr_component_count     = 0;
-	bool    cu_group_enable          = false;
-	bool    require_forward_progress = false;
-	bool    fp16_overflow            = false;
-};
-
-struct VsShaderResource2 {
-	bool    scratch_en        = false;
-	uint8_t user_sgpr         = 0;
-	bool    offchip_lds       = false;
-	bool    streamout_enabled = false;
-	uint8_t shared_vgprs      = 0;
-};
-
 struct HsShaderResource1 {
 	uint8_t vgprs                     = 0;
 	uint8_t priority                  = 0;
@@ -511,7 +430,6 @@ struct HsShaderResource2 {
 
 struct PsShaderResource1 {
 	uint8_t vgprs                    = 0;
-	uint8_t sgprs                    = 0;
 	uint8_t priority                 = 0;
 	uint8_t float_mode               = 0;
 	bool    dx10_clamp               = false;
@@ -535,7 +453,6 @@ struct PsStageRegisters {
 	uint64_t          data_addr = 0;
 	PsShaderResource1 rsrc1;
 	PsShaderResource2 rsrc2;
-	uint64_t          chksum = 0;
 };
 
 struct CsStageRegisters {
@@ -545,12 +462,13 @@ struct CsStageRegisters {
 	uint32_t num_thread_y              = 0;
 	uint32_t num_thread_z              = 0;
 	uint8_t  vgprs                     = 0;
-	uint8_t  sgprs                     = 0;
+	uint8_t  priority                  = 0;
 	uint8_t  float_mode                = 0;
 	bool     dx10_clamp                = false;
+	bool     debug_mode                = false;
 	bool     ieee_mode                 = false;
+	bool     require_forward_progress  = false;
 	bool     fp16_overflow             = false;
-	bool     bulky                     = false;
 	bool     threadgroup_configuration = false;
 	uint8_t  wave_size                 = 64;
 	bool     scratch_en                = false;
@@ -561,6 +479,7 @@ struct CsStageRegisters {
 	bool     tg_size_en                = false;
 	uint8_t  tidig_comp_cnt            = 0;
 	uint16_t lds_size                  = 0;
+	uint8_t  shared_vgprs              = 0;
 };
 
 struct EsStageRegisters {
@@ -568,21 +487,17 @@ struct EsStageRegisters {
 };
 
 struct LsStageRegisters {
-	uint64_t          data_addr = 0;
-	VsShaderResource1 rsrc1;
-	VsShaderResource2 rsrc2;
+	uint64_t data_addr = 0;
 };
 
 struct HsStageRegisters {
 	uint64_t          data_addr = 0;
 	HsShaderResource1 rsrc1;
 	HsShaderResource2 rsrc2;
-	uint64_t          chksum = 0;
 };
 
 struct GsShaderResource1 {
 	uint8_t vgprs                     = 0;
-	uint8_t sgprs                     = 0;
 	uint8_t priority                  = 0;
 	uint8_t float_mode                = 0;
 	bool    dx10_clamp                = false;
@@ -608,7 +523,6 @@ struct GsStageRegisters {
 	uint64_t          data_addr = 0;
 	GsShaderResource1 rsrc1;
 	GsShaderResource2 rsrc2;
-	uint64_t          chksum = 0;
 };
 
 struct ShaderRegisters {
@@ -664,8 +578,7 @@ struct ShaderRegisters {
 enum class UserSgprType { Unknown, Region, Vsharp };
 
 struct UserSgprInfo {
-	// Changed from 16 to 64
-	static constexpr int SGPRS_MAX = 64;
+	static constexpr int SGPRS_MAX = 32;
 
 	uint32_t     value[SGPRS_MAX] = {0};
 	UserSgprType type[SGPRS_MAX]  = {};
@@ -712,12 +625,6 @@ public:
 	void Reset() { *this = Context(); }
 
 	void SetColorBase(uint32_t slot, const ColorBase& base) { m_render_targets[slot].base = base; }
-	void SetColorPitch(uint32_t slot, const ColorPitch& pitch) {
-		m_render_targets[slot].pitch = pitch;
-	}
-	void SetColorSlice(uint32_t slot, const ColorSlice& slice) {
-		m_render_targets[slot].slice = slice;
-	}
 	void SetColorView(uint32_t slot, const ColorView& view) { m_render_targets[slot].view = view; }
 	void SetColorInfo(uint32_t slot, const ColorInfo& info) { m_render_targets[slot].info = info; }
 	void SetColorAttrib(uint32_t slot, const ColorAttrib& attrib) {
@@ -735,15 +642,9 @@ public:
 	void SetColorCmask(uint32_t slot, const ColorCmask& cmask) {
 		m_render_targets[slot].cmask = cmask;
 	}
-	void SetColorCmaskSlice(uint32_t slot, const ColorCmaskSlice& cmask_slice) {
-		m_render_targets[slot].cmask_slice = cmask_slice;
-	}
 	void SetColorFmask(uint32_t slot, const ColorFmask& fmask) {
 		m_render_targets[slot].fmask = fmask;
 	}
-	void SetColorFmaskSlice(uint32_t slot, const ColorFmaskSlice& fmask_slice) {
-		m_render_targets[slot].fmask_slice = fmask_slice;
-	}
 	void SetColorClearWord0(uint32_t slot, const ColorClearWord0& clear_word0) {
 		m_render_targets[slot].clear_word0 = clear_word0;
 	}
@@ -753,7 +654,6 @@ public:
 	void SetColorDccAddr(uint32_t slot, const ColorDccAddr& dcc_addr) {
 		m_render_targets[slot].dcc_addr = dcc_addr;
 	}
-	void SetColorSize(uint32_t slot, const ColorSize& size) { m_render_targets[slot].size = size; }
 	[[nodiscard]] const RenderTarget& GetRenderTarget(uint32_t slot) const {
 		return m_render_targets[slot];
 	}
@@ -1072,21 +972,16 @@ public:
 
 	void SetEsShaderBase(uint64_t addr) { m_vs.es_regs.data_addr = addr; }
 	void SetLsShaderBase(uint64_t addr) { m_vs.ls_regs.data_addr = addr; }
-	void SetLsShaderResource1(const VsShaderResource1& rsrc1) { m_vs.ls_regs.rsrc1 = rsrc1; }
-	void SetLsShaderResource2(const VsShaderResource2& rsrc2) { m_vs.ls_regs.rsrc2 = rsrc2; }
 	void SetHsShaderBase(uint64_t addr) { m_vs.hs_regs.data_addr = addr; }
 	void SetHsShaderResource1(const HsShaderResource1& rsrc1) { m_vs.hs_regs.rsrc1 = rsrc1; }
 	void SetHsShaderResource2(const HsShaderResource2& rsrc2) { m_vs.hs_regs.rsrc2 = rsrc2; }
-	void SetHsShaderChksum(uint32_t value) { m_vs.hs_regs.chksum = value; }
 	void SetGsShaderBase(uint64_t addr) { m_vs.gs_regs.data_addr = addr; }
 	void SetGsShaderResource1(const GsShaderResource1& rsrc1) { m_vs.gs_regs.rsrc1 = rsrc1; }
 	void SetGsShaderResource2(const GsShaderResource2& rsrc2) { m_vs.gs_regs.rsrc2 = rsrc2; }
-	void SetGsShaderChksum(uint32_t value) { m_vs.gs_regs.chksum = value; }
 
 	void SetPsShaderBase(uint64_t addr) { m_ps.ps_regs.data_addr = addr; }
 	void SetPsShaderResource1(const PsShaderResource1& rsrc1) { m_ps.ps_regs.rsrc1 = rsrc1; }
 	void SetPsShaderResource2(const PsShaderResource2& rsrc2) { m_ps.ps_regs.rsrc2 = rsrc2; }
-	void SetPsShaderChksum(uint32_t value) { m_ps.ps_regs.chksum = value; }
 
 	void SetCsShader(const CsStageRegisters& cs_regs) { m_cs.cs_regs = cs_regs; }
 	void SetCsWaveSize(uint8_t wave_size) { m_cs.cs_regs.wave_size = wave_size; }

src/graphics/guest_gpu/pm4.cpp

@@ -100,7 +100,7 @@ constexpr auto g_opcode_names   = MakeOpcodeNames();
 
 } // namespace
 
-void DumpPm4PacketStream(Common::File* file, uint32_t* cmd_buffer, uint32_t start_dw,
+void DumpPm4PacketStream(Common::File* file, const uint32_t* cmd_buffer, uint32_t start_dw,
                          uint32_t num_dw) {
 	// db_dump();
 
@@ -143,9 +143,9 @@ void DumpPm4PacketStream(Common::File* file, uint32_t* cmd_buffer, uint32_t star
 				if ((op == IT_SET_CONTEXT_REG_INDIRECT || op == IT_SET_SH_REG_INDIRECT ||
 				     op == IT_SET_UCONFIG_REG_INDIRECT) &&
 				    len == 4) {
-					auto* indirect_buffer =
-					    reinterpret_cast<uint32_t*>((static_cast<uint64_t>(cmd[0]) & 0xfffffffcu) |
-					                                (static_cast<uint64_t>(cmd[1]) << 32u));
+					auto* indirect_buffer = reinterpret_cast<const uint32_t*>(
+					    (static_cast<uint64_t>(cmd[0]) & 0xfffffffcu) |
+					    (static_cast<uint64_t>(cmd[1]) << 32u));
 					const uint32_t indirect_num_regs = cmd[3] & 0x3fffu;
 
 					for (uint32_t i = 0; indirect_buffer != nullptr && i < indirect_num_regs;

src/graphics/guest_gpu/pm4.h

@@ -98,6 +98,10 @@ constexpr uint32_t DB_RENDER_CONTROL_DEPTH_CLEAR_ENABLE_SHIFT       = 0;
 constexpr uint32_t DB_RENDER_CONTROL_DEPTH_CLEAR_ENABLE_MASK        = 0x1;
 constexpr uint32_t DB_RENDER_CONTROL_STENCIL_CLEAR_ENABLE_SHIFT     = 1;
 constexpr uint32_t DB_RENDER_CONTROL_STENCIL_CLEAR_ENABLE_MASK      = 0x1;
+constexpr uint32_t DB_RENDER_CONTROL_COPY_DEPTH_TO_COLOR_SHIFT      = 2;
+constexpr uint32_t DB_RENDER_CONTROL_COPY_DEPTH_TO_COLOR_MASK       = 0x1;
+constexpr uint32_t DB_RENDER_CONTROL_COPY_STENCIL_TO_COLOR_SHIFT    = 3;
+constexpr uint32_t DB_RENDER_CONTROL_COPY_STENCIL_TO_COLOR_MASK     = 0x1;
 constexpr uint32_t DB_RENDER_CONTROL_RESUMMARIZE_ENABLE_SHIFT       = 4;
 constexpr uint32_t DB_RENDER_CONTROL_RESUMMARIZE_ENABLE_MASK        = 0x1;
 constexpr uint32_t DB_RENDER_CONTROL_STENCIL_COMPRESS_DISABLE_SHIFT = 5;
@@ -163,22 +167,6 @@ constexpr uint32_t PA_SC_SCREEN_SCISSOR_BR_BR_Y_MASK  = 0xFFFF;
 
 constexpr uint32_t DB_DFSM_CONTROL = 0xE;
 
-constexpr uint32_t DB_DEPTH_INFO                          = 0xF;
-constexpr uint32_t DB_DEPTH_INFO_ADDR5_SWIZZLE_MASK_SHIFT = 0;
-constexpr uint32_t DB_DEPTH_INFO_ADDR5_SWIZZLE_MASK_MASK  = 0xF;
-constexpr uint32_t DB_DEPTH_INFO_ARRAY_MODE_SHIFT         = 4;
-constexpr uint32_t DB_DEPTH_INFO_ARRAY_MODE_MASK          = 0xF;
-constexpr uint32_t DB_DEPTH_INFO_PIPE_CONFIG_SHIFT        = 8;
-constexpr uint32_t DB_DEPTH_INFO_PIPE_CONFIG_MASK         = 0x1F;
-constexpr uint32_t DB_DEPTH_INFO_BANK_WIDTH_SHIFT         = 13;
-constexpr uint32_t DB_DEPTH_INFO_BANK_WIDTH_MASK          = 0x3;
-constexpr uint32_t DB_DEPTH_INFO_BANK_HEIGHT_SHIFT        = 15;
-constexpr uint32_t DB_DEPTH_INFO_BANK_HEIGHT_MASK         = 0x3;
-constexpr uint32_t DB_DEPTH_INFO_MACRO_TILE_ASPECT_SHIFT  = 17;
-constexpr uint32_t DB_DEPTH_INFO_MACRO_TILE_ASPECT_MASK   = 0x3;
-constexpr uint32_t DB_DEPTH_INFO_NUM_BANKS_SHIFT          = 19;
-constexpr uint32_t DB_DEPTH_INFO_NUM_BANKS_MASK           = 0x3;
-
 constexpr uint32_t DB_Z_INFO                               = 0x10;
 constexpr uint32_t DB_Z_INFO_FORMAT_SHIFT                  = 0;
 constexpr uint32_t DB_Z_INFO_FORMAT_MASK                   = 0x3;
@@ -222,16 +210,6 @@ constexpr uint32_t DB_STENCIL_READ_BASE  = 0x13;
 constexpr uint32_t DB_Z_WRITE_BASE       = 0x14;
 constexpr uint32_t DB_STENCIL_WRITE_BASE = 0x15;
 
-constexpr uint32_t DB_DEPTH_SIZE                       = 0x16;
-constexpr uint32_t DB_DEPTH_SIZE_PITCH_TILE_MAX_SHIFT  = 0;
-constexpr uint32_t DB_DEPTH_SIZE_PITCH_TILE_MAX_MASK   = 0x7FF;
-constexpr uint32_t DB_DEPTH_SIZE_HEIGHT_TILE_MAX_SHIFT = 11;
-constexpr uint32_t DB_DEPTH_SIZE_HEIGHT_TILE_MAX_MASK  = 0x7FF;
-
-constexpr uint32_t DB_DEPTH_SLICE                      = 0x17;
-constexpr uint32_t DB_DEPTH_SLICE_SLICE_TILE_MAX_SHIFT = 0;
-constexpr uint32_t DB_DEPTH_SLICE_SLICE_TILE_MAX_MASK  = 0x3FFFFF;
-
 constexpr uint32_t DB_Z_READ_BASE_HI                  = 0x1A;
 constexpr uint32_t DB_STENCIL_READ_BASE_HI            = 0x1B;
 constexpr uint32_t DB_Z_WRITE_BASE_HI                 = 0x1C;
@@ -424,11 +402,6 @@ constexpr uint32_t DB_DEPTH_CONTROL_STENCILFUNC_SHIFT                        = 8
 constexpr uint32_t DB_DEPTH_CONTROL_STENCILFUNC_MASK                         = 0x7;
 constexpr uint32_t DB_DEPTH_CONTROL_STENCILFUNC_BF_SHIFT                     = 20;
 constexpr uint32_t DB_DEPTH_CONTROL_STENCILFUNC_BF_MASK                      = 0x7;
-constexpr uint32_t DB_DEPTH_CONTROL_ENABLE_COLOR_WRITES_ON_DEPTH_FAIL_SHIFT  = 30;
-constexpr uint32_t DB_DEPTH_CONTROL_ENABLE_COLOR_WRITES_ON_DEPTH_FAIL_MASK   = 0x1;
-constexpr uint32_t DB_DEPTH_CONTROL_DISABLE_COLOR_WRITES_ON_DEPTH_PASS_SHIFT = 31;
-constexpr uint32_t DB_DEPTH_CONTROL_DISABLE_COLOR_WRITES_ON_DEPTH_PASS_MASK  = 0x1;
-
 constexpr uint32_t DB_EQAA                                  = 0x201;
 constexpr uint32_t DB_EQAA_MAX_ANCHOR_SAMPLES_SHIFT         = 0;
 constexpr uint32_t DB_EQAA_MAX_ANCHOR_SAMPLES_MASK          = 0x7;
@@ -560,21 +533,9 @@ constexpr uint32_t VGT_DRAW_PAYLOAD_CNTL  = 0x2A6;
 constexpr uint32_t VGT_ESGS_RING_ITEMSIZE = 0x2AB;
 constexpr uint32_t VGT_REUSE_OFF          = 0x2AD;
 
-constexpr uint32_t DB_HTILE_SURFACE                               = 0x2AF;
-constexpr uint32_t DB_HTILE_SURFACE_LINEAR_SHIFT                  = 0;
-constexpr uint32_t DB_HTILE_SURFACE_LINEAR_MASK                   = 0x1;
-constexpr uint32_t DB_HTILE_SURFACE_FULL_CACHE_SHIFT              = 1;
-constexpr uint32_t DB_HTILE_SURFACE_FULL_CACHE_MASK               = 0x1;
-constexpr uint32_t DB_HTILE_SURFACE_HTILE_USES_PRELOAD_WIN_SHIFT  = 2;
-constexpr uint32_t DB_HTILE_SURFACE_HTILE_USES_PRELOAD_WIN_MASK   = 0x1;
-constexpr uint32_t DB_HTILE_SURFACE_PRELOAD_SHIFT                 = 3;
-constexpr uint32_t DB_HTILE_SURFACE_PRELOAD_MASK                  = 0x1;
-constexpr uint32_t DB_HTILE_SURFACE_PREFETCH_WIDTH_SHIFT          = 4;
-constexpr uint32_t DB_HTILE_SURFACE_PREFETCH_WIDTH_MASK           = 0x3F;
-constexpr uint32_t DB_HTILE_SURFACE_PREFETCH_HEIGHT_SHIFT         = 10;
-constexpr uint32_t DB_HTILE_SURFACE_PREFETCH_HEIGHT_MASK          = 0x3F;
-constexpr uint32_t DB_HTILE_SURFACE_DST_OUTSIDE_ZERO_TO_ONE_SHIFT = 16;
-constexpr uint32_t DB_HTILE_SURFACE_DST_OUTSIDE_ZERO_TO_ONE_MASK  = 0x1;
+constexpr uint32_t DB_SHADING_RATE_ENCODING                             = 0x2AF;
+constexpr uint32_t DB_SHADING_RATE_ENCODING_SHADING_RATE_ENCODING_SHIFT = 19;
+constexpr uint32_t DB_SHADING_RATE_ENCODING_SHADING_RATE_ENCODING_MASK  = 0x3;
 
 constexpr uint32_t DB_SRESULTS_COMPARE_STATE0                          = 0x2B0;
 constexpr uint32_t DB_SRESULTS_COMPARE_STATE1                          = 0x2B1;
@@ -621,7 +582,6 @@ constexpr uint32_t PA_SC_SHADER_CONTROL                  = 0x310;
 constexpr uint32_t PA_SC_BINNER_CNTL_0                   = 0x311;
 constexpr uint32_t PA_SC_BINNER_CNTL_1                   = 0x312;
 constexpr uint32_t PA_SC_CONSERVATIVE_RASTERIZATION_CNTL = 0x313;
-constexpr uint32_t PA_SC_NGG_MODE_CNTL                   = 0x314;
 constexpr uint32_t CB_COLOR0_BASE                        = 0x318;
 
 constexpr uint32_t CB_COLOR0_VIEW                   = 0x31B;
@@ -649,24 +609,13 @@ constexpr uint32_t CB_COLOR0_INFO_BLEND_BYPASS_SHIFT              = 16;
 constexpr uint32_t CB_COLOR0_INFO_BLEND_BYPASS_MASK               = 0x1;
 constexpr uint32_t CB_COLOR0_INFO_ROUND_MODE_SHIFT                = 18;
 constexpr uint32_t CB_COLOR0_INFO_ROUND_MODE_MASK                 = 0x1;
-constexpr uint32_t CB_COLOR0_INFO_CMASK_IS_LINEAR_SHIFT           = 19;
-constexpr uint32_t CB_COLOR0_INFO_CMASK_IS_LINEAR_MASK            = 0x1;
 constexpr uint32_t CB_COLOR0_INFO_FMASK_COMPRESSION_DISABLE_SHIFT = 26;
 constexpr uint32_t CB_COLOR0_INFO_FMASK_COMPRESSION_DISABLE_MASK  = 0x1;
 constexpr uint32_t CB_COLOR0_INFO_FMASK_COMPRESS_1FRAG_ONLY_SHIFT = 27;
 constexpr uint32_t CB_COLOR0_INFO_FMASK_COMPRESS_1FRAG_ONLY_MASK  = 0x1;
 constexpr uint32_t CB_COLOR0_INFO_DCC_ENABLE_SHIFT                = 28;
 constexpr uint32_t CB_COLOR0_INFO_DCC_ENABLE_MASK                 = 0x1;
-constexpr uint32_t CB_COLOR0_INFO_CMASK_ADDR_TYPE_SHIFT           = 29;
-constexpr uint32_t CB_COLOR0_INFO_CMASK_ADDR_TYPE_MASK            = 0x3;
-constexpr uint32_t CB_COLOR0_INFO_ALT_TILE_MODE_SHIFT             = 31;
-constexpr uint32_t CB_COLOR0_INFO_ALT_TILE_MODE_MASK              = 0x1;
-
 constexpr uint32_t CB_COLOR0_ATTRIB                             = 0x31D;
-constexpr uint32_t CB_COLOR0_ATTRIB_TILE_MODE_INDEX_SHIFT       = 0;
-constexpr uint32_t CB_COLOR0_ATTRIB_TILE_MODE_INDEX_MASK        = 0x1F;
-constexpr uint32_t CB_COLOR0_ATTRIB_FMASK_TILE_MODE_INDEX_SHIFT = 5;
-constexpr uint32_t CB_COLOR0_ATTRIB_FMASK_TILE_MODE_INDEX_MASK  = 0x1F;
 constexpr uint32_t CB_COLOR0_ATTRIB_NUM_SAMPLES_SHIFT           = 12;
 constexpr uint32_t CB_COLOR0_ATTRIB_NUM_SAMPLES_MASK            = 0x7;
 constexpr uint32_t CB_COLOR0_ATTRIB_NUM_FRAGMENTS_SHIFT         = 15;
@@ -681,8 +630,6 @@ constexpr uint32_t CB_COLOR0_DCC_CONTROL_KEY_CLEAR_ENABLE_SHIFT
 constexpr uint32_t CB_COLOR0_DCC_CONTROL_KEY_CLEAR_ENABLE_MASK                  = 0x1;
 constexpr uint32_t CB_COLOR0_DCC_CONTROL_MAX_UNCOMPRESSED_BLOCK_SIZE_SHIFT      = 2;
 constexpr uint32_t CB_COLOR0_DCC_CONTROL_MAX_UNCOMPRESSED_BLOCK_SIZE_MASK       = 0x3;
-constexpr uint32_t CB_COLOR0_DCC_CONTROL_MIN_COMPRESSED_BLOCK_SIZE_SHIFT        = 4;
-constexpr uint32_t CB_COLOR0_DCC_CONTROL_MIN_COMPRESSED_BLOCK_SIZE_MASK         = 0x1;
 constexpr uint32_t CB_COLOR0_DCC_CONTROL_MAX_COMPRESSED_BLOCK_SIZE_SHIFT        = 5;
 constexpr uint32_t CB_COLOR0_DCC_CONTROL_MAX_COMPRESSED_BLOCK_SIZE_MASK         = 0x3;
 constexpr uint32_t CB_COLOR0_DCC_CONTROL_COLOR_TRANSFORM_SHIFT                  = 7;
@@ -739,31 +686,25 @@ constexpr uint32_t CB_COLOR0_ATTRIB3_CMASK_PIPE_ALIGNED_SHIFT = 26;
 constexpr uint32_t CB_COLOR0_ATTRIB3_CMASK_PIPE_ALIGNED_MASK  = 0x1;
 constexpr uint32_t CB_COLOR0_ATTRIB3_DCC_PIPE_ALIGNED_SHIFT   = 30;
 constexpr uint32_t CB_COLOR0_ATTRIB3_DCC_PIPE_ALIGNED_MASK    = 0x1;
+constexpr uint32_t CB_COLOR0_ATTRIB3_WRITE_VRS_RATE_HINT_TO_CMASK_SHIFT = 31;
+constexpr uint32_t CB_COLOR0_ATTRIB3_WRITE_VRS_RATE_HINT_TO_CMASK_MASK  = 0x1;
 
 constexpr uint32_t CB_COLOR7_ATTRIB3 = 0x3BF;
 
-/* Fake codes. Don't exist on real HW */
-
-constexpr uint32_t CX_NOP = 0x800003FF;
+constexpr uint32_t CX_NOP = 0xDB;
 
 constexpr uint32_t CX_NUM = 0x3FF + 1;
 
 /* Shader registers */
 
-constexpr uint32_t SPI_SHADER_TBA_LO_PS     = 0x0;
-constexpr uint32_t SPI_SHADER_PGM_RSRC4_PS  = 0x1;
-constexpr uint32_t SPI_SHADER_TMA_LO_PS     = 0x2;
-constexpr uint32_t SPI_SHADER_TMA_HI_PS     = 0x3;
-constexpr uint32_t SPI_SHADER_PGM_CHKSUM_PS = 0x6;
-constexpr uint32_t SPI_SHADER_PGM_RSRC3_PS  = 0x7;
-constexpr uint32_t SPI_SHADER_PGM_LO_PS     = 0x8;
-constexpr uint32_t SPI_SHADER_PGM_HI_PS     = 0x9;
+constexpr uint32_t SPI_SHADER_PACE_ID_PS            = 0x6;
+constexpr uint32_t SPI_GRAPHICS_SHADER_CONTROL_PS    = 0x7;
+constexpr uint32_t SPI_SHADER_PGM_LO_PS              = 0x8;
+constexpr uint32_t SPI_SHADER_PGM_HI_PS              = 0x9;
 
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_PS                        = 0xA;
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_PS_VGPRS_SHIFT            = 0;
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_PS_VGPRS_MASK             = 0x3F;
-constexpr uint32_t SPI_SHADER_PGM_RSRC1_PS_SGPRS_SHIFT            = 6;
-constexpr uint32_t SPI_SHADER_PGM_RSRC1_PS_SGPRS_MASK             = 0xF;
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_PS_PRIORITY_SHIFT         = 10;
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_PS_PRIORITY_MASK          = 0x3;
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_PS_FLOAT_MODE_SHIFT       = 12;
@@ -800,14 +741,11 @@ constexpr uint32_t SPI_SHADER_PGM_RSRC2_PS_SHARED_VGPR_CNT_MASK         = 0xF;
 constexpr uint32_t SPI_SHADER_USER_DATA_PS_0  = 0xC;
 constexpr uint32_t SPI_SHADER_USER_DATA_PS_15 = 0x1B;
 constexpr uint32_t SPI_SHADER_USER_DATA_PS_31 = 0x2B;
-constexpr uint32_t SPI_SHADER_REQ_CTRL_PS     = 0x30;
 constexpr uint32_t SPI_SHADER_USER_ACCUM_PS_0 = 0x32;
 constexpr uint32_t SPI_SHADER_USER_ACCUM_PS_3 = 0x35;
 
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_VS_VGPRS_SHIFT           = 0;
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_VS_VGPRS_MASK            = 0x3F;
-constexpr uint32_t SPI_SHADER_PGM_RSRC1_VS_SGPRS_SHIFT           = 6;
-constexpr uint32_t SPI_SHADER_PGM_RSRC1_VS_SGPRS_MASK            = 0xF;
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_VS_PRIORITY_SHIFT        = 10;
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_VS_PRIORITY_MASK         = 0x3;
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_VS_FLOAT_MODE_SHIFT      = 12;
@@ -838,19 +776,20 @@ constexpr uint32_t SPI_SHADER_PGM_RSRC2_VS_USER_SGPR_MSB_MASK    = 0x1;
 constexpr uint32_t SPI_SHADER_PGM_RSRC2_VS_SHARED_VGPR_CNT_SHIFT = 28;
 constexpr uint32_t SPI_SHADER_PGM_RSRC2_VS_SHARED_VGPR_CNT_MASK  = 0xF;
 
-constexpr uint32_t SPI_SHADER_PGM_CHKSUM_GS        = 0x80;
+constexpr uint32_t SPI_SHADER_PACE_ID_GS           = 0x80;
+// AGC compiler records use this numeric tag as opaque metadata. It is not a shader checksum
+// register and intentionally has no checksum-state consumer in HardwareContext.
+constexpr uint32_t SPI_SHADER_PGM_CHKSUM_GS        = SPI_SHADER_PACE_ID_GS;
 constexpr uint32_t SPI_SHADER_PGM_RSRC4_GS         = 0x81;
 constexpr uint32_t SPI_SHADER_USER_DATA_ADDR_LO_GS = 0x82;
 constexpr uint32_t SPI_SHADER_USER_DATA_ADDR_HI_GS = 0x83;
-constexpr uint32_t SPI_SHADER_PGM_RSRC3_GS         = 0x87;
+constexpr uint32_t SPI_GRAPHICS_SHADER_CONTROL_GS  = 0x87;
 constexpr uint32_t SPI_SHADER_PGM_LO_GS            = 0x88;
 constexpr uint32_t SPI_SHADER_PGM_HI_GS            = 0x89;
 
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_GS                        = 0x8A;
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_GS_VGPRS_SHIFT            = 0;
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_GS_VGPRS_MASK             = 0x3F;
-constexpr uint32_t SPI_SHADER_PGM_RSRC1_GS_SGPRS_SHIFT            = 6;
-constexpr uint32_t SPI_SHADER_PGM_RSRC1_GS_SGPRS_MASK             = 0xF;
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_GS_PRIORITY_SHIFT         = 10;
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_GS_PRIORITY_MASK          = 0x3;
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_GS_FLOAT_MODE_SHIFT       = 12;
@@ -891,22 +830,15 @@ constexpr uint32_t SPI_SHADER_PGM_RSRC2_GS_SHARED_VGPR_CNT_MASK   = 0xF;
 constexpr uint32_t SPI_SHADER_USER_DATA_GS_0                      = 0x8C;
 constexpr uint32_t SPI_SHADER_USER_DATA_GS_15                     = 0x9B;
 constexpr uint32_t SPI_SHADER_USER_DATA_GS_31                     = 0xAB;
-constexpr uint32_t SPI_SHADER_REQ_CTRL_ESGS                       = 0xB0;
 constexpr uint32_t SPI_SHADER_USER_ACCUM_ESGS_0                   = 0xB2;
 constexpr uint32_t SPI_SHADER_USER_ACCUM_ESGS_3                   = 0xB5;
-constexpr uint32_t SPI_SHADER_PGM_RSRC2_ES_GS                     = 0xBC;
 constexpr uint32_t SPI_SHADER_PGM_LO_ES                           = 0xC8;
 constexpr uint32_t SPI_SHADER_PGM_HI_ES                           = 0xC9;
-constexpr uint32_t SPI_SHADER_PGM_RSRC1_ES                        = 0xCA;
-constexpr uint32_t SPI_SHADER_PGM_RSRC2_ES                        = 0xCB;
-constexpr uint32_t SPI_SHADER_USER_DATA_ES_0                      = 0xCC;
-constexpr uint32_t SPI_SHADER_USER_DATA_ES_15                     = 0xDB;
-constexpr uint32_t SPI_SHADER_USER_DATA_ES_31                     = 0xEB;
+// Retained only as an opaque AGC compiler-record tag; 0x100 is not a PS5 SH register.
 constexpr uint32_t SPI_SHADER_PGM_CHKSUM_HS                       = 0x100;
-constexpr uint32_t SPI_SHADER_PGM_RSRC4_HS                        = 0x101;
 constexpr uint32_t SPI_SHADER_USER_DATA_ADDR_LO_HS                = 0x102;
 constexpr uint32_t SPI_SHADER_USER_DATA_ADDR_HI_HS                = 0x103;
-constexpr uint32_t SPI_SHADER_PGM_RSRC3_HS                        = 0x107;
+constexpr uint32_t SPI_GRAPHICS_SHADER_CONTROL_HS                 = 0x107;
 constexpr uint32_t SPI_SHADER_PGM_LO_HS                           = 0x108;
 constexpr uint32_t SPI_SHADER_PGM_HI_HS                           = 0x109;
 constexpr uint32_t SPI_SHADER_PGM_RSRC1_HS                        = 0x10A;
@@ -944,13 +876,10 @@ constexpr uint32_t SPI_SHADER_PGM_RSRC2_HS_SHARED_VGPR_CNT_MASK   = 0xF;
 constexpr uint32_t SPI_SHADER_USER_DATA_HS_0                      = 0x10C;
 constexpr uint32_t SPI_SHADER_USER_DATA_HS_15                     = 0x11B;
 constexpr uint32_t SPI_SHADER_USER_DATA_HS_31                     = 0x12B;
-constexpr uint32_t SPI_SHADER_REQ_CTRL_LSHS                       = 0x130;
 constexpr uint32_t SPI_SHADER_USER_ACCUM_LSHS_0                   = 0x132;
 constexpr uint32_t SPI_SHADER_USER_ACCUM_LSHS_3                   = 0x135;
 constexpr uint32_t SPI_SHADER_PGM_LO_LS                           = 0x148;
 constexpr uint32_t SPI_SHADER_PGM_HI_LS                           = 0x149;
-constexpr uint32_t SPI_SHADER_PGM_RSRC1_LS                        = 0x14A;
-constexpr uint32_t SPI_SHADER_PGM_RSRC2_LS                        = 0x14B;
 constexpr uint32_t COMPUTE_START_X                                = 0x204;
 constexpr uint32_t COMPUTE_START_Y                                = 0x205;
 constexpr uint32_t COMPUTE_START_Z                                = 0x206;
@@ -959,28 +888,26 @@ constexpr uint32_t COMPUTE_NUM_THREAD_Y                           = 0x208;
 constexpr uint32_t COMPUTE_NUM_THREAD_Z                           = 0x209;
 constexpr uint32_t COMPUTE_PGM_LO                                 = 0x20C;
 constexpr uint32_t COMPUTE_PGM_HI                                 = 0x20D;
-constexpr uint32_t COMPUTE_TBA_LO                                 = 0x20E;
-constexpr uint32_t COMPUTE_TBA_HI                                 = 0x20F;
-constexpr uint32_t COMPUTE_TMA_LO                                 = 0x210;
-constexpr uint32_t COMPUTE_TMA_HI                                 = 0x211;
 
 constexpr uint32_t COMPUTE_PGM_RSRC1                  = 0x212;
 constexpr uint32_t COMPUTE_PGM_RSRC1_VGPRS_SHIFT      = 0;
 constexpr uint32_t COMPUTE_PGM_RSRC1_VGPRS_MASK       = 0x3F;
-constexpr uint32_t COMPUTE_PGM_RSRC1_SGPRS_SHIFT      = 6;
-constexpr uint32_t COMPUTE_PGM_RSRC1_SGPRS_MASK       = 0xF;
+constexpr uint32_t COMPUTE_PGM_RSRC1_PRIORITY_SHIFT   = 10;
+constexpr uint32_t COMPUTE_PGM_RSRC1_PRIORITY_MASK    = 0x3;
 constexpr uint32_t COMPUTE_PGM_RSRC1_FLOAT_MODE_SHIFT = 12;
 constexpr uint32_t COMPUTE_PGM_RSRC1_FLOAT_MODE_MASK  = 0xFF;
 constexpr uint32_t COMPUTE_PGM_RSRC1_DX10_CLAMP_SHIFT = 21;
 constexpr uint32_t COMPUTE_PGM_RSRC1_DX10_CLAMP_MASK  = 0x1;
+constexpr uint32_t COMPUTE_PGM_RSRC1_DEBUG_MODE_SHIFT = 22;
+constexpr uint32_t COMPUTE_PGM_RSRC1_DEBUG_MODE_MASK  = 0x1;
 constexpr uint32_t COMPUTE_PGM_RSRC1_IEEE_MODE_SHIFT  = 23;
 constexpr uint32_t COMPUTE_PGM_RSRC1_IEEE_MODE_MASK   = 0x1;
-constexpr uint32_t COMPUTE_PGM_RSRC1_BULKY_SHIFT      = 24;
-constexpr uint32_t COMPUTE_PGM_RSRC1_BULKY_MASK       = 0x1;
 constexpr uint32_t COMPUTE_PGM_RSRC1_FP16_OVFL_SHIFT  = 26;
 constexpr uint32_t COMPUTE_PGM_RSRC1_FP16_OVFL_MASK   = 0x1;
 constexpr uint32_t COMPUTE_PGM_RSRC1_WGP_MODE_SHIFT   = 29;
 constexpr uint32_t COMPUTE_PGM_RSRC1_WGP_MODE_MASK    = 0x1;
+constexpr uint32_t COMPUTE_PGM_RSRC1_FWD_PROGRESS_SHIFT = 31;
+constexpr uint32_t COMPUTE_PGM_RSRC1_FWD_PROGRESS_MASK  = 0x1;
 
 constexpr uint32_t COMPUTE_DISPATCH_INITIATOR_CS_W32_EN_SHIFT = 15;
 constexpr uint32_t COMPUTE_DISPATCH_INITIATOR_CS_W32_EN_MASK  = 0x1;
@@ -1011,21 +938,16 @@ constexpr uint32_t COMPUTE_PGM_RSRC2_LDS_SIZE_SHIFT       = 15;
 constexpr uint32_t COMPUTE_PGM_RSRC2_LDS_SIZE_MASK        = 0x1FF;
 
 constexpr uint32_t COMPUTE_RESOURCE_LIMITS    = 0x215;
-constexpr uint32_t COMPUTE_DESTINATION_EN_SE0 = 0x216;
-constexpr uint32_t COMPUTE_DESTINATION_EN_SE1 = 0x217;
 constexpr uint32_t COMPUTE_TMPRING_SIZE       = 0x218;
-constexpr uint32_t COMPUTE_DESTINATION_EN_SE2 = 0x219;
-constexpr uint32_t COMPUTE_DESTINATION_EN_SE3 = 0x21A;
 constexpr uint32_t COMPUTE_USER_ACCUM_0       = 0x224;
 constexpr uint32_t COMPUTE_PGM_RSRC3          = 0x228;
-constexpr uint32_t COMPUTE_SHADER_CHKSUM      = 0x22A;
+constexpr uint32_t COMPUTE_PGM_RSRC3_SHARED_VGPRS_SHIFT = 0;
+constexpr uint32_t COMPUTE_PGM_RSRC3_SHARED_VGPRS_MASK  = 0xF;
+constexpr uint32_t COMPUTE_PACE_ID             = 0x22A;
 constexpr uint32_t COMPUTE_USER_DATA_0        = 0x240;
 constexpr uint32_t COMPUTE_USER_DATA_15       = 0x24F;
-constexpr uint32_t COMPUTE_DISPATCH_TUNNEL    = 0x27D;
 
-/* Fake codes. Don't exist on real HW */
-
-constexpr uint32_t SH_NOP = 0x800002FF;
+constexpr uint32_t SH_NOP = 0x280;
 
 constexpr uint32_t SH_NUM = 0x2FF + 1;
 
@@ -1042,6 +964,7 @@ constexpr uint32_t GE_INDX_OFFSET            = 0x24A;
 constexpr uint32_t GE_MULTI_PRIM_IB_RESET_EN = 0x24B;
 constexpr uint32_t VGT_HS_OFFCHIP_PARAM      = 0x24F;
 constexpr uint32_t VGT_TF_MEMORY_BASE        = 0x250;
+constexpr uint32_t IA_MULTI_VGT_PARAM        = 0x258;
 
 constexpr uint32_t GE_CNTL                     = 0x25B;
 constexpr uint32_t GE_CNTL_PRIM_GRP_SIZE_SHIFT = 0;
@@ -1063,32 +986,19 @@ constexpr uint32_t GE_USER_VGPR_EN_EN_USER_VGPR2_MASK  = 0x1;
 constexpr uint32_t GE_USER_VGPR_EN_EN_USER_VGPR3_SHIFT = 2;
 constexpr uint32_t GE_USER_VGPR_EN_EN_USER_VGPR3_MASK  = 0x1;
 
-constexpr uint32_t IA_MULTI_VGT_PARAM       = 0x258;
 constexpr uint32_t TA_CS_BC_BASE_ADDR       = 0x380;
 constexpr uint32_t TA_CS_BC_BASE_ADDR_HI    = 0x381;
-constexpr uint32_t TEXTURE_GRADIENT_FACTORS = 0x382;
+constexpr uint32_t TEXTURE_GRADIENT_FACTORS = 0x383;
+constexpr uint32_t TEXTURE_GRADIENT_CONTROL = 0x384;
 constexpr uint32_t GDS_OA_CNTL              = 0x41D;
 constexpr uint32_t GDS_OA_COUNTER           = 0x41E;
 constexpr uint32_t GDS_OA_ADDRESS           = 0x41F;
 
-/* Fake codes. Don't exist on real HW */
-
-constexpr uint32_t FSR_EXTEND_SUBPIXEL_ROUNDING = 0x80003FF4;
-constexpr uint32_t FSR_ALPHA_VALUE0             = 0x80003FF5;
-constexpr uint32_t FSR_ALPHA_VALUE1             = 0x80003FF6;
-constexpr uint32_t FSR_CONTROL_POINT0           = 0x80003FF7;
-constexpr uint32_t FSR_CONTROL_POINT1           = 0x80003FF8;
-constexpr uint32_t FSR_CONTROL_POINT2           = 0x80003FF9;
-constexpr uint32_t FSR_CONTROL_POINT3           = 0x80003FFA;
-constexpr uint32_t FSR_WINDOW0                  = 0x80003FFB;
-constexpr uint32_t FSR_WINDOW1                  = 0x80003FFC;
-constexpr uint32_t TEXTURE_GRADIENT_CONTROL     = 0x80003FFD;
-constexpr uint32_t MEMORY_MAPPING_MASK          = 0x80003FFE;
-constexpr uint32_t UC_NOP                       = 0x80003FFF;
+constexpr uint32_t UC_NOP = 0xA2;
 
 constexpr uint32_t UC_NUM = 0x3FFF + 1;
 
-void DumpPm4PacketStream(Common::File* file, uint32_t* cmd_buffer, uint32_t start_dw,
+void DumpPm4PacketStream(Common::File* file, const uint32_t* cmd_buffer, uint32_t start_dw,
                          uint32_t num_dw);
 
 } // namespace Libs::Graphics::Pm4

src/graphics/host_gpu/graphicContext.h

@@ -3,20 +3,88 @@
 
 #include "common/abi.h"
 #include "common/common.h"
+#include "common/threads.h"
 #include "graphics/host_gpu/vulkanCommon.h" // IWYU pragma: export
-#include "graphics/host_gpu/vulkanInstance.h"
 
+#include <map>
+#include <mutex>
+#include <tuple>
 #include <vector>
 #include <vk_mem_alloc.h>
 
 namespace Libs::Graphics {
 
-class CommandBuffer;
 struct VulkanBuffer;
 struct VulkanImage;
 struct VulkanMemory;
 
-struct GraphicContext: public VulkanInstance {
+inline constexpr uint32_t VULKAN_TARGET_API_VERSION = VK_API_VERSION_1_3;
+
+struct GraphicContext {
+	vk::Instance                       instance                              = nullptr;
+	vk::DebugUtilsMessengerEXT         debug_messenger                       = nullptr;
+	vk::PhysicalDevice                 physical_device                       = nullptr;
+	vk::PhysicalDeviceProperties       physical_device_properties            = {};
+	vk::PhysicalDeviceMemoryProperties physical_device_memory_properties     = {};
+	vk::Device                         device                                = nullptr;
+	VmaAllocator                       allocator                             = nullptr;
+	bool                               memory_budget_ext_enabled             = false;
+	bool                               rt_extensions_enabled                 = false;
+	bool                               compute_subgroup_size_control_enabled = false;
+	bool                               compute_wave64_supported              = false;
+	bool                               sample_rate_shading_enabled           = false;
+	uint32_t                           subgroup_size                         = 0;
+	uint32_t                           min_subgroup_size                     = 0;
+	uint32_t                           max_subgroup_size                     = 0;
+	uint32_t                           max_push_descriptors                  = 0;
+	vk::ShaderStageFlags               required_subgroup_size_stages         = {};
+	Common::Mutex                      queue_mutex;
+	uint32_t                           queue_family = static_cast<uint32_t>(-1);
+	vk::Queue                          queue        = nullptr;
+
+	[[nodiscard]] const vk::PhysicalDeviceProperties& GetPhysicalDeviceProperties() const {
+		return physical_device_properties;
+	}
+
+	[[nodiscard]] const vk::PhysicalDeviceMemoryProperties&
+	GetPhysicalDeviceMemoryProperties() const {
+		return physical_device_memory_properties;
+	}
+
+	[[nodiscard]] vk::FormatProperties GetFormatProperties(vk::Format format) const {
+		std::scoped_lock lock(m_format_properties_mutex);
+		auto [it, inserted] = m_format_properties.try_emplace(format);
+		if (inserted) {
+			physical_device.getFormatProperties(format, &it->second);
+		}
+		return it->second;
+	}
+
+	[[nodiscard]] vk::Result GetImageFormatProperties(vk::Format format, vk::ImageType type,
+	                                                  vk::ImageTiling            tiling,
+	                                                  vk::ImageUsageFlags        usage,
+	                                                  vk::ImageCreateFlags       flags,
+	                                                  vk::ImageFormatProperties* properties) const {
+		using Key = std::tuple<vk::Format, vk::ImageType, vk::ImageTiling, vk::ImageUsageFlags,
+		                       vk::ImageCreateFlags>;
+		std::scoped_lock lock(m_image_format_properties_mutex);
+		auto [it, inserted] =
+		    m_image_format_properties.try_emplace(Key {format, type, tiling, usage, flags});
+		if (inserted) {
+			it->second.first = physical_device.getImageFormatProperties(format, type, tiling, usage,
+			                                                            flags, &it->second.second);
+		}
+		if (properties != nullptr) {
+			*properties = it->second.second;
+		}
+		return it->second.first;
+	}
+
+	[[nodiscard]] vk::DeviceSize StorageMinAlignment() const {
+		const auto alignment = physical_device_properties.limits.minStorageBufferOffsetAlignment;
+		return alignment != 0 ? alignment : 1;
+	}
+
 	[[nodiscard]] bool CreateAllocator();
 	void               DestroyAllocator();
 	void               LogMemoryBudget() const;
@@ -24,7 +92,6 @@ struct GraphicContext: public VulkanInstance {
 	[[nodiscard]] uint64_t GetDeviceMemoryUsage() const;
 	[[nodiscard]] uint64_t GetTotalMemoryBudget() const;
 	void                   CreateBuffer(uint64_t size, VulkanBuffer& buffer);
-	void                   DeleteBuffer(VulkanBuffer& buffer);
 	[[nodiscard]] bool     CreateImage(const vk::ImageCreateInfo& info, VulkanImage& image);
 	void                   DeleteImage(VulkanImage& image);
 	void                   MapMemory(VulkanMemory& memory, void*& data);
@@ -36,6 +103,15 @@ struct GraphicContext: public VulkanInstance {
 
 	uint32_t screen_width  = 0;
 	uint32_t screen_height = 0;
+
+private:
+	mutable std::mutex                                 m_format_properties_mutex;
+	mutable std::map<vk::Format, vk::FormatProperties> m_format_properties;
+	mutable std::mutex                                 m_image_format_properties_mutex;
+	mutable std::map<std::tuple<vk::Format, vk::ImageType, vk::ImageTiling, vk::ImageUsageFlags,
+	                            vk::ImageCreateFlags>,
+	                 std::pair<vk::Result, vk::ImageFormatProperties>>
+	    m_image_format_properties;
 };
 
 struct VulkanMemory {

src/graphics/host_gpu/memoryTracker.cpp

@@ -18,8 +18,8 @@ MemoryTracker::~MemoryTracker() = default;
 #if KYTY_BUILD == KYTY_BUILD_DEBUG
 void MemoryTracker::ValidateGpuDirtyPages(const RangeSet& dirty, uint64_t vaddr, uint64_t size,
                                           const char* operation) const noexcept {
-	if (vaddr == 0 || size == 0 || size > UINT64_MAX - vaddr ||
-	    (vaddr & (TRACKER_PAGE_SIZE - 1)) != 0 || (size & (TRACKER_PAGE_SIZE - 1)) != 0) {
+	if (!GuestRange {vaddr, size}.Valid() || (vaddr & (TRACKER_PAGE_SIZE - 1)) != 0 ||
+	    (size & (TRACKER_PAGE_SIZE - 1)) != 0) {
 		EXIT("MemoryTracker: invalid dirty-page validation range\n");
 	}
 	for (auto page = vaddr; page < vaddr + size; page += TRACKER_PAGE_SIZE) {
@@ -37,9 +37,6 @@ void MemoryTracker::ValidateGpuDirtyPages(const RangeSet& dirty, uint64_t vaddr,
 void MemoryTracker::ValidateGpuDirtyOwnership(const RangeSet& dirty, uint64_t vaddr, uint64_t size,
                                               const char* operation) {
 	ValidateRange(vaddr, size);
-	if (vaddr + size > UINT64_MAX - (TRACKER_PAGE_SIZE - 1)) {
-		EXIT("MemoryTracker: dirty ownership range alignment overflow\n");
-	}
 	const auto begin = vaddr & ~(TRACKER_PAGE_SIZE - 1);
 	const auto end   = (vaddr + size + TRACKER_PAGE_SIZE - 1) & ~(TRACKER_PAGE_SIZE - 1);
 	for (auto page = begin; page < end; page += TRACKER_PAGE_SIZE) {
@@ -56,8 +53,7 @@ void MemoryTracker::ValidateGpuDirtyOwnership(const RangeSet& dirty, uint64_t va
 #endif
 
 void MemoryTracker::ValidateRange(uint64_t vaddr, uint64_t size) {
-	if (vaddr == 0 || size == 0 || vaddr >= TRACKER_ADDRESS_SIZE ||
-	    size > TRACKER_ADDRESS_SIZE - vaddr) {
+	if (!GuestRange {vaddr, size}.Valid()) {
 		EXIT("invalid memory tracker range\n");
 	}
 }

src/graphics/host_gpu/pageManager.cpp

@@ -110,7 +110,7 @@ private:
 };
 
 void ValidateRange(uint64_t vaddr, uint64_t size) {
-	if (vaddr == 0 || size == 0 || vaddr >= ADDRESS_SIZE || size > ADDRESS_SIZE - vaddr) {
+	if (!GuestRange {vaddr, size}.Valid()) {
 		Fatal("invalid range vaddr=0x%016" PRIx64 ", size=0x%016" PRIx64, vaddr, size);
 	}
 }

src/graphics/host_gpu/rangeSet.h

@@ -53,12 +53,30 @@ public:
 		}
 	}
 
+	void Clear() { m_ranges.clear(); }
+
+	template <typename Func>
+	void ForEach(Func&& func) const {
+		for (const auto& [begin, end]: m_ranges) {
+			func(begin, end);
+		}
+	}
+
 	[[nodiscard]] std::vector<Range> Intersections(uint64_t address, uint64_t size) const {
 		std::vector<Range> result;
 		ForEachIntersection(address, size, [&result](Range range) { result.push_back(range); });
 		return result;
 	}
 
+	[[nodiscard]] bool Intersects(uint64_t address, uint64_t size) const {
+		const auto end = End(address, size);
+		auto       it  = m_ranges.lower_bound(address);
+		if (it != m_ranges.begin() && std::prev(it)->second > address) {
+			return true;
+		}
+		return it != m_ranges.end() && it->first < end;
+	}
+
 	[[nodiscard]] bool Contains(uint64_t address, uint64_t size) const {
 		const auto end = End(address, size);
 		auto       it  = m_ranges.upper_bound(address);
@@ -89,7 +107,7 @@ public:
 
 private:
 	static uint64_t End(uint64_t address, uint64_t size) {
-		if (address == 0 || size == 0 || size > UINT64_MAX - address) {
+		if (size == 0 || size > UINT64_MAX - address) {
 			EXIT("invalid range-set address or size\n");
 		}
 		return address + size;

src/graphics/host_gpu/regionDefinitions.h

@@ -4,6 +4,8 @@
 #include "common/bitArray.h"
 #include "common/common.h"
 
+#include <compare>
+
 namespace Libs::Graphics {
 
 constexpr uint64_t TRACKER_PAGE_SIZE    = 4ull * 1024ull;
@@ -11,6 +13,20 @@ constexpr uint64_t TRACKER_REGION_SIZE  = 4ull * 1024ull * 1024ull;
 constexpr uint64_t TRACKER_ADDRESS_SIZE = 1ull << 40u;
 constexpr size_t   TRACKER_REGION_PAGES = TRACKER_REGION_SIZE / TRACKER_PAGE_SIZE;
 
+struct GuestRange {
+	uint64_t address = 0;
+	uint64_t size    = 0;
+
+	[[nodiscard]] constexpr bool Empty() const noexcept { return address == 0 && size == 0; }
+	[[nodiscard]] constexpr bool Valid() const noexcept {
+		return address != 0 && size != 0 && address < TRACKER_ADDRESS_SIZE &&
+		       size <= TRACKER_ADDRESS_SIZE - address;
+	}
+	[[nodiscard]] constexpr bool     ValidOrEmpty() const noexcept { return Empty() || Valid(); }
+	[[nodiscard]] constexpr uint64_t End() const noexcept { return address + size; }
+	auto                             operator<=>(const GuestRange&) const = default;
+};
+
 enum class DirtySource { Cpu, Gpu };
 using RegionBits = Common::BitArray<TRACKER_REGION_PAGES>;
 static_assert(sizeof(RegionBits) == TRACKER_REGION_PAGES / 8);

src/graphics/host_gpu/renderer/cache/bufferCache.h

@@ -3,12 +3,15 @@
 
 #include "common/abi.h"
 #include "common/common.h"
+#include "common/lruCache.h"
+#include "common/slotVector.h"
 #include "graphics/host_gpu/memoryTracker.h"
 #include "graphics/host_gpu/rangeSet.h"
+#include "graphics/host_gpu/renderer/cache/faultManager.h"
+#include "graphics/host_gpu/renderer/cache/multiLevelPageTable.h"
 #include "graphics/host_gpu/renderer/cache/streamBuffer.h"
 
 #include <map>
-#include <memory>
 #include <span>
 #include <utility>
 #include <vector>
@@ -16,93 +19,95 @@
 namespace Libs::Graphics {
 
 struct GraphicContext;
-class CommandBuffer;
 class CommandScheduler;
 class TextureCache;
 
-struct BufferBinding {
-	std::shared_ptr<void> owner;
-	vk::Buffer            buffer = nullptr;
-	uint64_t              offset = 0;
-};
-
-struct ImageBufferSource {
-	Buffer*  buffer = nullptr;
-	uint64_t offset = 0;
-};
+using BufferId = Common::SlotId;
+inline constexpr BufferId NULL_BUFFER_ID {0};
 
 class BufferCache {
 public:
-	static constexpr uint64_t CACHING_PAGE_SIZE = 16ull * 1024ull;
-	static constexpr uint64_t GetBufferOffset(uint64_t vaddr) {
-		return vaddr & (CACHING_PAGE_SIZE - 1);
-	}
+	static constexpr uint32_t CACHING_PAGEBITS  = 14;
+	static constexpr uint64_t CACHING_PAGESIZE  = uint64_t {1} << CACHING_PAGEBITS;
+	static constexpr uint64_t CACHING_NUMPAGES  = uint64_t {1} << (40 - CACHING_PAGEBITS);
+	static constexpr uint64_t BDA_PAGETABLE_SIZE =
+	    CACHING_NUMPAGES * sizeof(vk::DeviceAddress);
 
 	BufferCache(GraphicContext& graphics, CommandScheduler& scheduler, PageManager& page_manager,
 	            TextureCache& texture_cache);
 	~BufferCache();
 	KYTY_CLASS_NO_COPY(BufferCache);
 
-	void                        InvalidateMemory(uint64_t vaddr, uint64_t size);
-	void                        ReadMemory(uint64_t vaddr, uint64_t size, bool is_write = false);
-	void                        UnmapMemory(uint64_t vaddr, uint64_t size);
-	[[nodiscard]] BufferBinding ObtainBuffer(CommandBuffer& command, uint64_t vaddr, uint64_t size,
-	                                         bool is_written = false, bool is_read = true,
-	                                         bool is_formatted = false);
-	[[nodiscard]] StreamBuffer& GetUtilityBuffer(MemoryUsage usage) noexcept;
-	[[nodiscard]] Buffer&       GetGdsBuffer() noexcept { return m_gds_buffer; }
-	[[nodiscard]] const Buffer& GetGdsBuffer() const noexcept { return m_gds_buffer; }
-	[[nodiscard]] BufferBinding UploadTransient(const void* data, uint64_t size,
-	                                            uint64_t alignment);
-	[[nodiscard]] std::shared_ptr<Buffer> ObtainNullBuffer();
-	[[nodiscard]] ImageBufferSource       ObtainBufferForImage(uint64_t vaddr, uint64_t size);
-	void FillBuffer(uint64_t vaddr, uint64_t size, uint32_t value, bool is_gds = false);
-	void CopyBuffer(uint64_t dst_vaddr, uint64_t src_vaddr, uint64_t size, bool dst_gds = false,
-	                bool src_gds = false);
+	void                   InvalidateMemory(uint64_t vaddr, uint64_t size);
+	void                   ReadMemory(uint64_t vaddr, uint64_t size, bool is_write = false);
+	[[nodiscard]] Buffer&  GetBuffer(BufferId id) { return m_slot_buffers[id]; }
+	[[nodiscard]] BufferId FindBuffer(uint64_t vaddr, uint64_t size);
+	[[nodiscard]] std::pair<Buffer*, uint64_t> ObtainBuffer(uint64_t vaddr, uint64_t size,
+	                                                        bool     is_written,
+	                                                        bool     is_texel_buffer = false,
+	                                                        BufferId id              = {});
+	[[nodiscard]] StreamBuffer&                GetUtilityBuffer(MemoryUsage usage) noexcept {
+		switch (usage) {
+			case MemoryUsage::Upload: return m_staging_buffer;
+			case MemoryUsage::Stream: return m_stream_buffer;
+			case MemoryUsage::Download: return m_download_buffer;
+			case MemoryUsage::DeviceLocal: return m_device_buffer;
+		}
+		EXIT("BufferCache: invalid utility-buffer usage\n");
+	}
+	[[nodiscard]] const Buffer* GetGdsBuffer() const noexcept { return &m_gds_buffer; }
+	[[nodiscard]] Buffer* GetBdaPageTableBuffer() noexcept { return &m_bda_pagetable_buffer; }
+	[[nodiscard]] Buffer* GetFaultBuffer() noexcept { return m_fault_manager.GetFaultBuffer(); }
+	[[nodiscard]] std::pair<Buffer*, uint64_t> ObtainBufferForImage(uint64_t vaddr, uint64_t size);
+	void FillBuffer(uint64_t vaddr, uint64_t size, uint32_t value, bool is_gds);
+	void CopyBuffer(uint64_t dst_vaddr, uint64_t src_vaddr, uint64_t size, bool dst_gds,
+	                bool src_gds);
 	// Cache-index and exact dirty-range queries require GPU-thread serialization.
 	[[nodiscard]] bool IsRegionRegistered(uint64_t vaddr, uint64_t size);
 	[[nodiscard]] bool HasGpuDirtyBytes(uint64_t vaddr, uint64_t size);
 	[[nodiscard]] bool IsRegionCpuModified(uint64_t vaddr, uint64_t size);
 	[[nodiscard]] bool IsRegionGpuModified(uint64_t vaddr, uint64_t size);
+	void               ProcessFaultBuffer();
+	void               SynchronizeBuffersInRange(uint64_t vaddr, uint64_t size);
 	void               RunGarbageCollector();
 
 private:
 	friend struct BufferCacheTestAccess;
 
-	struct CacheRange {
-		uint64_t address = 0;
-		uint64_t size    = 0;
-	};
-	struct CachedBuffer;
 	struct DownloadCopy;
-	struct DownloadRange;
-	struct RetiredBuffer;
+	using PageTable = MultiLevelPageTable<BufferId, CACHING_PAGEBITS, 40, 16>;
+	static_assert(CACHING_PAGESIZE == (uint64_t {1} << PageTable::kPageBits));
 	static constexpr uint64_t               DOWNLOAD_ALIGNMENT = 64;
-	[[nodiscard]] static uint64_t           AlignDown(uint64_t value) noexcept;
-	[[nodiscard]] static uint64_t           AlignUp(uint64_t value);
 	[[nodiscard]] static constexpr uint64_t AlignDownload(uint64_t size) noexcept {
 		return (size + DOWNLOAD_ALIGNMENT - 1) & ~(DOWNLOAD_ALIGNMENT - 1);
 	}
 	[[nodiscard]] static std::pair<uint64_t, uint64_t> DownloadEnvelope(const DownloadCopy& copy);
-	[[nodiscard]] static bool ResolveOverlap(CacheRange& merged, CacheRange candidate) noexcept;
-	void Upload(CommandBuffer& command, Buffer& destination, uint64_t destination_offset,
-	            const void* source, uint64_t size);
-	[[nodiscard]] CachedBuffer& GetOrCreateBuffer(CommandBuffer& command, uint64_t vaddr,
-	                                              uint64_t size);
-	[[nodiscard]] bool SynchronizeBuffer(CommandBuffer& command, Buffer& buffer, uint64_t vaddr,
-	                                     uint64_t size, bool is_written, bool is_texel_buffer);
+	void WriteDataBuffer(Buffer& buffer, uint64_t address, const void* source, uint64_t size);
+	void TouchBuffer(const Buffer& buffer);
+	[[nodiscard]] BufferId CreateBuffer(uint64_t vaddr, uint64_t size);
+	void                   Register(BufferId id);
+	void Unregister(BufferId id);
+	template <bool insert>
+	void ChangeRegister(BufferId id);
+	void DeleteBuffer(BufferId id);
+	[[nodiscard]] bool SynchronizeBuffer(Buffer& buffer, uint64_t vaddr, uint64_t size,
+	                                     bool is_written, bool is_texel_buffer);
+	[[nodiscard]] vk::Buffer UploadCopies(Buffer& buffer, std::span<vk::BufferCopy> copies,
+	                                      uint64_t total_size);
 	[[nodiscard]] bool SynchronizeBufferFromImage(Buffer& buffer, uint64_t vaddr, uint64_t size);
-	[[nodiscard]] std::vector<DownloadRange> RecordDownloads(std::span<const DownloadCopy> copies);
-	void PublishDownloads(std::span<const DownloadRange> downloads);
-	void QueueGarbageDownload(std::span<const DownloadCopy> copies, RetiredBuffer retire);
+	void DownloadBufferMemory(std::span<const DownloadCopy> copies);
 	void WriteHostMemory(uint64_t vaddr, std::span<const uint8_t> data);
 	void ReadMemoryOnGpu(uint64_t vaddr, uint64_t size, bool is_write);
 
 	GraphicContext&                                   m_graphics;
 	CommandScheduler&                                 m_scheduler;
+	FaultManager                                      m_fault_manager;
 	Buffer                                            m_gds_buffer;
-	std::shared_ptr<Buffer>                           m_null_buffer;
-	std::map<uint64_t, std::unique_ptr<CachedBuffer>> m_buffers;
+	Buffer                                            m_bda_pagetable_buffer;
+	Common::SlotVector<Buffer>                        m_slot_buffers;
+	Common::LeastRecentlyUsedCache<BufferId, uint64_t> m_lru_cache;
+	std::map<uint64_t, BufferId>                      m_buffers;
+	PageTable                                         m_page_table;
 	RangeSet                                          m_gpu_modified_ranges;
 	MemoryTracker                                     m_memory_tracker;
 	StreamBuffer                                      m_staging_buffer;
@@ -110,7 +115,7 @@ private:
 	StreamBuffer                                      m_download_buffer;
 	StreamBuffer                                      m_device_buffer;
 	TextureCache&                                     m_texture_cache;
-	uint64_t                                          m_total_used_memory = 0;
+	uint64_t                                          m_total_used_memory  = 0;
 	uint64_t m_trigger_gc_memory  = 1ull * 1024 * 1024 * 1024;
 	uint64_t m_critical_gc_memory = 2ull * 1024 * 1024 * 1024;
 	uint64_t m_gc_tick            = 0;

src/graphics/host_gpu/renderer/cache/faultManager.cpp

@@ -0,0 +1,167 @@
+#include "graphics/host_gpu/renderer/cache/faultManager.h"
+
+#include "common/assert.h"
+#include "common/logging/log.h"
+#include "gpu_tiler_shaders/fault_buffer_process_spv.h"
+#include "graphics/host_gpu/graphicContext.h"
+#include "graphics/host_gpu/renderer/cache/bufferCache.h"
+#include "graphics/host_gpu/renderer/commandScheduler.h"
+#include "graphics/host_gpu/vulkanCommon.h"
+
+#include <bit>
+#include <cinttypes>
+#include <cstring>
+#include <limits>
+
+namespace Libs::Graphics {
+
+namespace {
+
+constexpr size_t MaxPageFaults    = 1024;
+constexpr size_t PageFaultAreaSize = MaxPageFaults * sizeof(uint64_t);
+
+} // namespace
+
+FaultManager::FaultManager(GraphicContext& graphics, CommandScheduler& scheduler,
+                           BufferCache& buffer_cache, uint32_t caching_pagebits,
+                           uint64_t caching_num_pages)
+    : m_graphics(graphics), m_scheduler(scheduler), m_buffer_cache(buffer_cache),
+      m_caching_pagesize(uint64_t {1} << caching_pagebits),
+      m_caching_num_pages(caching_num_pages), m_fault_buffer_size(caching_num_pages / 8),
+      m_fault_buffer(graphics, scheduler, MemoryUsage::DeviceLocal, 0, AllFlags,
+                     m_fault_buffer_size),
+      m_download_buffer(graphics, scheduler, MemoryUsage::Download, 0, AllFlags,
+                        MaxPendingFaults * PageFaultAreaSize) {
+	SetVulkanObjectNameF(m_graphics.device, m_fault_buffer.Handle(), "Fault Buffer");
+
+	const vk::DescriptorSetLayoutBinding bindings[] {
+	    {0, vk::DescriptorType::eStorageBuffer, 1, vk::ShaderStageFlagBits::eCompute, nullptr},
+	    {1, vk::DescriptorType::eStorageBuffer, 1, vk::ShaderStageFlagBits::eCompute, nullptr},
+	};
+	vk::DescriptorSetLayoutCreateInfo layout_info {};
+	layout_info.sType        = vk::StructureType::eDescriptorSetLayoutCreateInfo;
+	layout_info.flags        = vk::DescriptorSetLayoutCreateFlagBits::ePushDescriptorKHR;
+	layout_info.bindingCount = std::size(bindings);
+	layout_info.pBindings    = bindings;
+	RequireVulkanSuccess(
+	    m_graphics.device.createDescriptorSetLayout(&layout_info, nullptr,
+	                                                &m_fault_process_desc_layout),
+	    "create fault-buffer descriptor layout");
+
+	vk::ShaderModuleCreateInfo module_info {};
+	module_info.sType    = vk::StructureType::eShaderModuleCreateInfo;
+	module_info.codeSize = std::size(FAULT_BUFFER_PROCESS_SPV) * sizeof(uint32_t);
+	module_info.pCode    = FAULT_BUFFER_PROCESS_SPV;
+	vk::ShaderModule module = nullptr;
+	RequireVulkanSuccess(m_graphics.device.createShaderModule(&module_info, nullptr, &module),
+	                     "create fault-buffer shader module");
+
+	vk::PipelineLayoutCreateInfo pipeline_layout_info {};
+	pipeline_layout_info.sType          = vk::StructureType::ePipelineLayoutCreateInfo;
+	pipeline_layout_info.setLayoutCount = 1;
+	pipeline_layout_info.pSetLayouts    = &m_fault_process_desc_layout;
+	RequireVulkanSuccess(
+	    m_graphics.device.createPipelineLayout(&pipeline_layout_info, nullptr,
+	                                           &m_fault_process_pipeline_layout),
+	    "create fault-buffer pipeline layout");
+
+	vk::PipelineShaderStageCreateInfo stage {};
+	stage.sType  = vk::StructureType::ePipelineShaderStageCreateInfo;
+	stage.stage  = vk::ShaderStageFlagBits::eCompute;
+	stage.module = module;
+	stage.pName  = "main";
+	vk::ComputePipelineCreateInfo pipeline_info {};
+	pipeline_info.sType  = vk::StructureType::eComputePipelineCreateInfo;
+	pipeline_info.stage  = stage;
+	pipeline_info.layout = m_fault_process_pipeline_layout;
+	const auto result = m_graphics.device.createComputePipelines(
+	    nullptr, 1, &pipeline_info, nullptr, &m_fault_process_pipeline);
+	m_graphics.device.destroyShaderModule(module, nullptr);
+	RequireVulkanSuccess(result, "create fault-buffer pipeline");
+	SetVulkanObjectNameF(m_graphics.device, m_fault_process_pipeline, "Fault Buffer Parser");
+}
+
+FaultManager::~FaultManager() {
+	m_graphics.device.destroyPipeline(m_fault_process_pipeline, nullptr);
+	m_graphics.device.destroyPipelineLayout(m_fault_process_pipeline_layout, nullptr);
+	m_graphics.device.destroyDescriptorSetLayout(m_fault_process_desc_layout, nullptr);
+}
+
+void FaultManager::ProcessFaultBuffer() {
+	if (const auto wait_tick = m_fault_areas[m_current_area]; wait_tick != 0) {
+		m_scheduler.Wait(wait_tick);
+		m_scheduler.PopPendingOperations();
+	}
+
+	const auto offset = m_current_area * PageFaultAreaSize;
+	auto*      mapped = m_download_buffer.Mapped().data() + offset;
+	std::memset(mapped, 0, PageFaultAreaSize);
+	m_download_buffer.Flush(offset, PageFaultAreaSize);
+
+	vk::BufferMemoryBarrier2 pre_barrier {};
+	pre_barrier.sType         = vk::StructureType::eBufferMemoryBarrier2;
+	pre_barrier.srcStageMask  = vk::PipelineStageFlagBits2::eAllCommands;
+	pre_barrier.srcAccessMask = vk::AccessFlagBits2::eShaderWrite;
+	pre_barrier.dstStageMask  = vk::PipelineStageFlagBits2::eComputeShader;
+	pre_barrier.dstAccessMask = vk::AccessFlagBits2::eShaderRead;
+	pre_barrier.buffer        = m_fault_buffer.Handle();
+	pre_barrier.offset        = 0;
+	pre_barrier.size          = m_fault_buffer_size;
+	auto post_barrier         = pre_barrier;
+	post_barrier.srcStageMask  = vk::PipelineStageFlagBits2::eComputeShader;
+	post_barrier.srcAccessMask = vk::AccessFlagBits2::eShaderWrite;
+	post_barrier.dstStageMask  = vk::PipelineStageFlagBits2::eAllCommands;
+	post_barrier.dstAccessMask = vk::AccessFlagBits2::eShaderWrite;
+
+	const vk::DescriptorBufferInfo infos[] {
+	    {m_fault_buffer.Handle(), 0, m_fault_buffer_size},
+	    {m_download_buffer.Handle(), offset, PageFaultAreaSize},
+	};
+	std::array<vk::WriteDescriptorSet, 2> writes {};
+	for (uint32_t index = 0; index < writes.size(); ++index) {
+		writes[index].sType           = vk::StructureType::eWriteDescriptorSet;
+		writes[index].dstBinding      = index;
+		writes[index].descriptorCount = 1;
+		writes[index].descriptorType  = vk::DescriptorType::eStorageBuffer;
+		writes[index].pBufferInfo     = &infos[index];
+	}
+
+	m_scheduler.EndRendering();
+	auto command = m_scheduler.Current().Handle();
+	vk::DependencyInfo dependency {};
+	dependency.sType                    = vk::StructureType::eDependencyInfo;
+	dependency.dependencyFlags          = vk::DependencyFlagBits::eByRegion;
+	dependency.bufferMemoryBarrierCount = 1;
+	dependency.pBufferMemoryBarriers    = &pre_barrier;
+	command.pipelineBarrier2(dependency);
+	command.bindPipeline(vk::PipelineBindPoint::eCompute, m_fault_process_pipeline);
+	command.pushDescriptorSetKHR(vk::PipelineBindPoint::eCompute,
+	                             m_fault_process_pipeline_layout, 0, writes);
+	const auto num_threads    = m_caching_num_pages / 32;
+	const auto num_workgroups = (num_threads + 63) / 64;
+	command.dispatch(static_cast<uint32_t>(num_workgroups), 1, 1);
+	dependency.pBufferMemoryBarriers = &post_barrier;
+	command.pipelineBarrier2(dependency);
+
+	const auto area = m_current_area;
+	m_scheduler.DeferOperation([this, mapped, offset, area] {
+		m_download_buffer.Invalidate(offset, PageFaultAreaSize);
+		m_fault_ranges.Clear();
+		const auto* faults = std::bit_cast<const uint64_t*>(mapped);
+		const auto  count  = static_cast<uint32_t>(faults[0]);
+		for (uint32_t index = 1; index <= count; ++index) {
+			m_fault_ranges.Add(faults[index], m_caching_pagesize);
+			LOGF("Accessed non-GPU cached memory at 0x%016" PRIx64 "\n", faults[index]);
+		}
+		m_fault_ranges.ForEach([this](uint64_t start, uint64_t end) {
+			EXIT_IF(end - start > std::numeric_limits<uint32_t>::max());
+			(void)m_buffer_cache.FindBuffer(start, end - start);
+		});
+		m_fault_areas[area] = 0;
+	});
+
+	m_fault_areas[m_current_area++] = m_scheduler.CurrentTick();
+	m_current_area %= MaxPendingFaults;
+}
+
+} // namespace Libs::Graphics

src/graphics/host_gpu/renderer/cache/faultManager.h

@@ -0,0 +1,46 @@
+#ifndef EMULATOR_SRC_GRAPHICS_HOST_GPU_RENDERER_CACHE_FAULTMANAGER_H_
+#define EMULATOR_SRC_GRAPHICS_HOST_GPU_RENDERER_CACHE_FAULTMANAGER_H_
+
+#include "common/abi.h"
+#include "graphics/host_gpu/rangeSet.h"
+#include "graphics/host_gpu/renderer/cache/streamBuffer.h"
+
+#include <array>
+#include <cstdint>
+
+namespace Libs::Graphics {
+
+class BufferCache;
+
+class FaultManager {
+	static constexpr size_t MaxPendingFaults = 8;
+
+public:
+	FaultManager(GraphicContext& graphics, CommandScheduler& scheduler, BufferCache& buffer_cache,
+	             uint32_t caching_pagebits, uint64_t caching_num_pages);
+	~FaultManager();
+	KYTY_CLASS_NO_COPY(FaultManager);
+
+	[[nodiscard]] Buffer* GetFaultBuffer() noexcept { return &m_fault_buffer; }
+	void                  ProcessFaultBuffer();
+
+private:
+	GraphicContext&                            m_graphics;
+	CommandScheduler&                          m_scheduler;
+	BufferCache&                               m_buffer_cache;
+	RangeSet                                   m_fault_ranges;
+	uint64_t                                   m_caching_pagesize;
+	uint64_t                                   m_caching_num_pages;
+	uint64_t                                   m_fault_buffer_size;
+	Buffer                                     m_fault_buffer;
+	Buffer                                     m_download_buffer;
+	std::array<uint64_t, MaxPendingFaults>      m_fault_areas {};
+	uint32_t                                   m_current_area = 0;
+	vk::DescriptorSetLayout                    m_fault_process_desc_layout = nullptr;
+	vk::Pipeline                               m_fault_process_pipeline = nullptr;
+	vk::PipelineLayout                         m_fault_process_pipeline_layout = nullptr;
+};
+
+} // namespace Libs::Graphics
+
+#endif // EMULATOR_SRC_GRAPHICS_HOST_GPU_RENDERER_CACHE_FAULTMANAGER_H_

src/graphics/host_gpu/renderer/cache/gpuResourceManager.cpp

@@ -35,8 +35,7 @@ bool GpuResourceManager::InvalidateMemory(uint64_t vaddr, uint64_t size) {
 }
 
 bool GpuResourceManager::IsMapped(uint64_t vaddr, uint64_t size) const noexcept {
-	if (vaddr == 0 || size == 0 || vaddr >= TRACKER_ADDRESS_SIZE ||
-	    size > TRACKER_ADDRESS_SIZE - vaddr) {
+	if (!GuestRange {vaddr, size}.Valid()) {
 		return false;
 	}
 	std::shared_lock lock(m_mapped_ranges_mutex);
@@ -60,10 +59,10 @@ void GpuResourceManager::UnmapMemory(uint64_t vaddr, uint64_t size) {
 	const auto unmap = [this, vaddr, size] {
 		if (m_scheduler.Active()) {
 			const auto tick = m_scheduler.CurrentTick();
-			m_scheduler.FinishCurrent();
+			m_scheduler.Finish();
 			m_scheduler.WaitPriorityOperations(tick);
 		}
-		m_buffer_cache.UnmapMemory(vaddr, size);
+		m_buffer_cache.InvalidateMemory(vaddr, size);
 		m_texture_cache.UnmapMemory(vaddr, size);
 		m_page_manager.OnGpuUnmap(vaddr, size);
 		std::lock_guard lock(m_mapped_ranges_mutex);
@@ -76,7 +75,19 @@ void GpuResourceManager::UnmapMemory(uint64_t vaddr, uint64_t size) {
 	m_gpu->SendCommandSync(unmap);
 }
 
+void GpuResourceManager::PrepareBda() {
+	std::shared_lock lock(m_mapped_ranges_mutex);
+	m_mapped_ranges.ForEach([this](uint64_t start, uint64_t end) {
+		m_buffer_cache.SynchronizeBuffersInRange(start, end - start);
+	});
+	m_fault_process_pending = true;
+}
+
 void GpuResourceManager::RunGarbageCollector() {
+	if (m_fault_process_pending) {
+		m_fault_process_pending = false;
+		m_buffer_cache.ProcessFaultBuffer();
+	}
 	m_texture_cache.ProcessDownloadImages();
 	m_texture_cache.RunGarbageCollector();
 	m_buffer_cache.RunGarbageCollector();

src/graphics/host_gpu/renderer/cache/gpuResourceManager.h

@@ -30,6 +30,7 @@ public:
 	[[nodiscard]] bool IsMapped(uint64_t vaddr, uint64_t size) const noexcept;
 	void               MapMemory(uint64_t vaddr, uint64_t size);
 	void               UnmapMemory(uint64_t vaddr, uint64_t size);
+	void               PrepareBda();
 	void               RunGarbageCollector();
 
 private:
@@ -40,6 +41,7 @@ private:
 	mutable std::shared_mutex m_mapped_ranges_mutex;
 	RangeSet                  m_mapped_ranges;
 	GuestGpu*                 m_gpu = nullptr;
+	bool                      m_fault_process_pending = false;
 };
 
 } // namespace Libs::Graphics

src/graphics/host_gpu/renderer/cache/streamBuffer.cpp

@@ -73,8 +73,12 @@ Buffer::Buffer(GraphicContext& graphics, CommandScheduler& scheduler, MemoryUsag
 	buffer_info.usage       = flags;
 	buffer_info.sharingMode = vk::SharingMode::eExclusive;
 
+	const bool with_bda = bool(flags & vk::BufferUsageFlagBits::eShaderDeviceAddress);
+	const VmaAllocationCreateFlags bda_flag =
+	    with_bda ? VMA_ALLOCATION_CREATE_DEDICATED_MEMORY_BIT : 0;
 	VmaAllocationCreateInfo allocation_info {};
-	allocation_info.flags = VMA_ALLOCATION_CREATE_WITHIN_BUDGET_BIT | AllocationFlags(usage);
+	allocation_info.flags =
+	    VMA_ALLOCATION_CREATE_WITHIN_BUDGET_BIT | bda_flag | AllocationFlags(usage);
 	allocation_info.usage = AllocationUsage(usage);
 	allocation_info.preferredFlags = usage == MemoryUsage::DeviceLocal
 	                                     ? VkMemoryPropertyFlags {}
@@ -99,6 +103,13 @@ Buffer::Buffer(GraphicContext& graphics, CommandScheduler& scheduler, MemoryUsag
 	m_buffer->memory.type            = allocation_result.memoryType;
 	m_buffer->memory.unique_id       = VulkanNextMemoryUniqueId();
 	graphics.device.getBufferMemoryRequirements(m_buffer->buffer, &m_buffer->memory.requirements);
+	if (static_cast<bool>(flags & vk::BufferUsageFlagBits::eShaderDeviceAddress)) {
+		vk::BufferDeviceAddressInfo address_info {};
+		address_info.sType  = vk::StructureType::eBufferDeviceAddressInfo;
+		address_info.buffer = m_buffer->buffer;
+		m_device_address    = graphics.device.getBufferAddress(address_info);
+		EXIT_IF(m_device_address == 0);
+	}
 
 	VkMemoryPropertyFlags properties = 0;
 	vmaGetAllocationMemoryProperties(graphics.allocator, m_buffer->memory.allocation, &properties);
@@ -122,6 +133,11 @@ vk::Buffer Buffer::Handle() const noexcept {
 	return m_buffer->buffer;
 }
 
+vk::DeviceAddress Buffer::BufferDeviceAddress() const noexcept {
+	EXIT_IF(m_device_address == 0);
+	return m_device_address;
+}
+
 bool Buffer::IsInBounds(uint64_t address, uint64_t size) const noexcept {
 	return address >= m_cpu_address && size <= m_size && address - m_cpu_address <= m_size - size;
 }
@@ -141,6 +157,15 @@ void Buffer::Flush(uint64_t offset, uint64_t size) {
 	}
 }
 
+void Buffer::Invalidate(uint64_t offset, uint64_t size) {
+	EXIT_IF(m_usage != MemoryUsage::Download || offset > m_size || size > m_size - offset);
+	if (!m_is_coherent && size != 0) {
+		const auto result =
+		    vmaInvalidateAllocation(m_graphics->allocator, m_buffer->memory.allocation, offset, size);
+		EXIT_NOT_IMPLEMENTED(static_cast<vk::Result>(result) != vk::Result::eSuccess);
+	}
+}
+
 vk::BufferMemoryBarrier Buffer::Barrier(uint64_t offset, uint64_t size, vk::AccessFlags source,
                                         vk::AccessFlags destination) const {
 	if (Handle() == nullptr || size == 0 || offset > m_size || size > m_size - offset) {
@@ -310,16 +335,6 @@ void StreamBuffer::Commit() {
 	watch.tick        = tick;
 }
 
-void StreamBuffer::Invalidate(uint64_t offset, uint64_t size) {
-	EXIT_IF(Usage() != MemoryUsage::Download || offset > Size() || size > Size() - offset);
-	if (IsCoherent() || size == 0) {
-		return;
-	}
-	const auto result = vmaInvalidateAllocation(Graphics().allocator,
-	                                            NativeBuffer().memory.allocation, offset, size);
-	EXIT_NOT_IMPLEMENTED(static_cast<vk::Result>(result) != vk::Result::eSuccess);
-}
-
 uint64_t StreamBuffer::Copy(const void* source, uint64_t size, uint64_t alignment) {
 	EXIT_IF(source == nullptr);
 	const auto [data, offset] = Map(size, alignment);

src/graphics/host_gpu/renderer/cache/streamBuffer.h

@@ -48,12 +48,14 @@ public:
 	[[nodiscard]] bool               IsCoherent() const noexcept { return m_is_coherent; }
 	[[nodiscard]] MemoryUsage        Usage() const noexcept { return m_usage; }
 	[[nodiscard]] uint64_t           CpuAddress() const noexcept { return m_cpu_address; }
+	[[nodiscard]] vk::DeviceAddress BufferDeviceAddress() const noexcept;
 	[[nodiscard]] uint64_t           Offset(uint64_t address) const noexcept {
 		return address - m_cpu_address;
 	}
 	[[nodiscard]] bool IsInBounds(uint64_t address, uint64_t size) const noexcept;
 	void               Write(uint64_t offset, const void* source, uint64_t size);
 	void               Flush(uint64_t offset, uint64_t size);
+	void               Invalidate(uint64_t offset, uint64_t size);
 	void CopyFrom(CommandBuffer& command, const Buffer& source, uint64_t source_offset,
 	              uint64_t destination_offset, uint64_t size,
 	              vk::AccessFlags source_before      = vk::AccessFlagBits::eMemoryWrite,
@@ -65,6 +67,10 @@ public:
 	                                                   vk::AccessFlagBits::eMemoryWrite);
 	void Fill(uint64_t offset, uint64_t size, uint32_t value);
 
+	// BufferCache state lives directly on the resource.
+	bool   is_deleted = false;
+	size_t lru_id     = 0;
+
 protected:
 	[[nodiscard]] GraphicContext&   Graphics() const noexcept { return *m_graphics; }
 	[[nodiscard]] CommandScheduler& Scheduler() const noexcept { return *m_scheduler; }
@@ -80,6 +86,7 @@ private:
 	MemoryUsage                   m_usage       = MemoryUsage::DeviceLocal;
 	uint64_t                      m_cpu_address = 0;
 	uint64_t                      m_size        = 0;
+	vk::DeviceAddress             m_device_address = 0;
 	std::unique_ptr<VulkanBuffer> m_buffer;
 	std::span<uint8_t>            m_mapped;
 	bool                          m_is_coherent = false;
@@ -93,9 +100,6 @@ public:
 	[[nodiscard]] std::pair<uint8_t*, uint64_t> Map(uint64_t size, uint64_t alignment = 0,
 	                                                bool allow_wait = true);
 	void                                        Commit();
-	// Download mappings become visible to the CPU only after their GPU completion tick is free.
-	// Call this from the scheduler's deferred completion operation before reading Mapped().
-	void                   Invalidate(uint64_t offset, uint64_t size);
 	[[nodiscard]] uint64_t Copy(const void* source, uint64_t size, uint64_t alignment = 0);
 
 private:

src/graphics/host_gpu/renderer/cache/textureCache.h

@@ -4,17 +4,17 @@
 #include "common/abi.h"
 #include "common/common.h"
 #include "common/lruCache.h"
+#include "common/slotVector.h"
 #include "graphics/host_gpu/pageManager.h"
 #include "graphics/host_gpu/regionManager.h"
 #include "graphics/host_gpu/renderer/cache/multiLevelPageTable.h"
 #include "graphics/host_gpu/renderer/image/blitHelper.h"
 #include "graphics/host_gpu/renderer/image/image.h"
+#include "graphics/host_gpu/renderer/image/tiler.h"
 
-#include <compare>
 #include <map>
-#include <memory>
-#include <set>
-#include <utility>
+#include <unordered_map>
+#include <unordered_set>
 #include <vector>
 
 namespace Libs::Graphics {
@@ -25,8 +25,6 @@ class BufferCache;
 class CommandBuffer;
 class CommandScheduler;
 class RenderExecutor;
-class StreamBuffer;
-class TileManager;
 struct TextureCacheTestAccess;
 
 class TextureCache {
@@ -57,15 +55,17 @@ public:
 	[[nodiscard]] vk::ImageView FindTexture(ImageId id, const ImageDesc& desc);
 	[[nodiscard]] vk::ImageView FindRenderTarget(ImageId id, const ImageDesc& desc);
 	[[nodiscard]] vk::ImageView FindDepthTarget(ImageId id, const ImageDesc& desc);
-	[[nodiscard]] Image&        GetImage(ImageId id);
-	[[nodiscard]] const Image&  GetImage(ImageId id) const;
-	void                        MarkGpuWritten(ImageId id);
+	[[nodiscard]] Image&        GetImage(ImageId id) {
+		auto& image = m_slot_images[id];
+		TouchImage(image);
+		return image;
+	}
+	void MarkGpuWritten(ImageId id);
 
 	[[nodiscard]] bool ClearImageFromBuffer(CommandBuffer& command, uint64_t address, uint64_t size,
 	                                        uint32_t packed_clear);
 	void               InvalidateMemory(uint64_t address, uint64_t size);
-	[[nodiscard]] bool InvalidateMemoryFromGPU(uint64_t address, uint64_t size,
-	                                           bool formatted_buffer_write = false);
+	void               InvalidateMemoryFromGPU(uint64_t address, uint64_t size);
 	[[nodiscard]] RegionInfo QueryRegion(uint64_t address, uint64_t size);
 
 	[[nodiscard]] bool IsMeta(uint64_t address);
@@ -86,16 +86,11 @@ private:
 	struct ColorTransferPlan;
 	struct DownloadPlan;
 
-	struct Slot {
-		std::shared_ptr<Image> image;
-		uint32_t               generation = 1;
-	};
-
 	struct MetaDataInfo {
 		// A guest metadata-fill dispatch may initialize DCC before its render target is bound.
 		// PendingDcc retains that exact fill until FindRenderTarget classifies the address,
 		// without exposing an unconfirmed buffer address to the normal metadata heuristics.
-		// Keep all surface metadata in this one shadPS4-style entry so CMask/FMask can be
+		// Keep all surface metadata in one entry so CMask/FMask can be
 		// registered beside HTile and DCC without introducing parallel tracking paths.
 		enum class Type : uint8_t { PendingDcc, CMask, FMask, HTile, Dcc };
 
@@ -114,26 +109,21 @@ private:
 	using ImageIds       = InlinePageOwnerList<ImageId, 16>;
 	using ImagePageTable = MultiLevelPageTable<ImageIds, 20, 40, 10>;
 
-	[[nodiscard]] Image&                 ResolveImage(ImageId id);
-	[[nodiscard]] const Image&           ResolveImage(ImageId id) const;
-	[[nodiscard]] std::shared_ptr<Image> ResolveOwner(ImageId id) const;
-	[[nodiscard]] ImageId                InsertImage(const ImageInfo& info);
-	[[nodiscard]] ImageId                GetNullImage(const ImageDesc& desc);
-	void                                 RegisterImage(ImageId id);
-	void                                 UnregisterImage(ImageId id);
-	void                                 DeleteImage(ImageId id);
-	void                                 FreeImage(ImageId id);
-	void                                 RetainImage(CommandBuffer& command, ImageId id);
-	void                                 TouchImage(Image& image);
-	void                                 TrackImage(ImageId id);
-	void                                 TrackImageHead(ImageId id);
-	void                                 TrackImageTail(ImageId id);
-	void                                 UntrackImage(ImageId id);
-	void                                 UntrackImageHead(ImageId id);
-	void                                 UntrackImageTail(ImageId id);
-	void                                 MarkAsMaybeDirty(ImageId id, Image& image);
-	void                                 TrackImageDownload(ImageId id);
-	void                                 TrackImageDownloadLocked(ImageId id, Image& image);
+	[[nodiscard]] ImageId     InsertImage(const ImageInfo& info);
+	[[nodiscard]] ImageId     GetNullImage(const ImageDesc& desc);
+	void                      RegisterImage(ImageId id);
+	void                      UnregisterImage(ImageId id);
+	void                      DeleteImage(ImageId id);
+	void                      FreeImage(ImageId id);
+	void                      TouchImage(Image& image);
+	void                      TrackImage(ImageId id);
+	void                      TrackImageHead(ImageId id);
+	void                      TrackImageTail(ImageId id);
+	void                      UntrackImage(ImageId id);
+	void                      UntrackImageHead(ImageId id);
+	void                      UntrackImageTail(ImageId id);
+	void                      MarkAsMaybeDirty(ImageId id, Image& image);
+	void                      TrackImageDownload(ImageId id, Image& image);
 	[[nodiscard]] static bool SameBacking(const ImageInfo& cached, const ImageInfo& requested,
 	                                      bool exact_format);
 	[[nodiscard]] static BindingType UploadBinding(const Image& image);
@@ -162,31 +152,24 @@ private:
 	[[nodiscard]] bool CopyD16(Image& destination, Image& source);
 	void               CopyImage(ImageId destination, ImageId source);
 	void               AssociateStencil(ImageId depth, GuestRange stencil);
-	void               AssociateStencilLocked(ImageId depth, GuestRange stencil);
 	void CopyImageMip(ImageId destination, ImageId source, uint32_t mip, uint32_t layer);
 	void ValidateImageDesc(const ImageDesc& desc) const;
 
-	void InvalidateCpuAliases(uint64_t address, uint64_t size);
-	void ClearGpuModified(ImageId id);
-
-	void                                        DownloadImage(ImageId id);
-	[[nodiscard]] bool                          TryDownloadImage(ImageId id);
-	[[nodiscard]] std::pair<uint8_t*, uint64_t> MapDownload(uint64_t size, uint64_t alignment);
-	void QueueDownload(GuestRange range, StreamBuffer& download, uint8_t* mapped, uint64_t offset);
+	void               InvalidateCpuAliases(uint64_t address, uint64_t size);
+	[[nodiscard]] bool TryDownloadImage(ImageId id);
 
 	GraphicContext&                                   m_graphics;
 	CommandScheduler&                                 m_scheduler;
 	TrackingSpinLock                                  m_lock;
 	PageManager&                                      m_page_manager;
 	BlitHelper                                        m_blit_helper;
-	std::unique_ptr<TileManager>                      m_tiler;
+	TileManager                                       m_tiler;
 	BufferCache&                                      m_buffer_cache;
-	std::vector<Slot>                                 m_slots;
-	std::vector<uint32_t>                             m_free_slots;
+	Common::SlotVector<Image>                         m_slot_images;
 	ImagePageTable                                    m_image_page_table;
-	std::map<vk::Format, ImageId>                     m_null_images;
+	std::unordered_map<vk::Format, ImageId>           m_null_images;
 	Common::LeastRecentlyUsedCache<ImageId, uint64_t> m_lru_cache;
-	std::set<ImageId>                                 m_download_images;
+	std::unordered_set<ImageId>                       m_download_images;
 	std::map<uint64_t, MetaDataInfo>                  m_surface_metas;
 	uint64_t                                          m_total_used_memory  = 0;
 	uint64_t                                          m_trigger_gc_memory  = 0;

src/graphics/host_gpu/renderer/colorRenderTarget.cpp

@@ -9,12 +9,12 @@
 #include "graphics/host_gpu/graphicContext.h"
 #include "graphics/host_gpu/renderer/debug.h"
 #include "graphics/host_gpu/renderer/image/textureCommon.h"
-#include "graphics/host_gpu/renderer/pipeline/descriptorCache.h"
 #include "graphics/host_gpu/renderer/render.h"
 #include "graphics/host_gpu/renderer/renderContext.h"
 #include "graphics/host_gpu/vulkanCommon.h"
 
 #include <algorithm>
+#include <array>
 #include <atomic>
 
 namespace Libs::Graphics {
@@ -44,7 +44,7 @@ static void ResolveDccClearInfo(RenderColorInfo& info, vk::Format format, bool h
 }
 
 // NOLINTNEXTLINE(readability-function-cognitive-complexity)
-void RenderExecutor::ResolveRenderColorTarget(uint64_t submit_id, RenderCommandBuffer& buffer,
+void RenderExecutor::ResolveRenderColorTarget(uint64_t submit_id, CommandBuffer& buffer,
                                               RenderColorInfo& r,
                                               uint32_t         render_target_slice_offset,
                                               uint32_t render_target_slot, bool ignore_target_mask,
@@ -144,12 +144,23 @@ void RenderExecutor::ResolveRenderColorTarget(uint64_t submit_id, RenderCommandB
 	uint32_t   pitch  = 0;
 	uint64_t   size   = 0;
 	bool       tile   = false;
-	const bool volume = rt.attrib3.dimension == 2;
-	if (rt.attrib3.dimension != 1 && !volume) {
+	static constexpr std::array image_types {Prospero::ImageType::kColor1D,
+	                                         Prospero::ImageType::kColor2D,
+	                                         Prospero::ImageType::kColor3D};
+	if (rt.attrib3.dimension >= image_types.size()) {
 		EXIT("unsupported render-target dimension: %u\n", rt.attrib3.dimension);
 	}
+	const auto image_type = image_types[rt.attrib3.dimension];
+	const bool is_1d      = image_type == Prospero::ImageType::kColor1D;
+	const bool volume     = image_type == Prospero::ImageType::kColor3D;
+	if (is_1d && rt.attrib2.height != 0) {
+		EXIT("1D render target has nonzero height: %u\n", rt.attrib2.height);
+	}
 	if (!volume && rt.attrib3.depth != 0) {
-		EXIT("2D render target has nonzero depth: %u\n", rt.attrib3.depth);
+		EXIT("non-3D render target has nonzero depth: %u\n", rt.attrib3.depth);
+	}
+	if (is_1d && samples != 1) {
+		EXIT("multisampled 1D render targets are unsupported\n");
 	}
 	if (volume && samples != 1) {
 		EXIT("multisampled 3D render targets are unsupported\n");
@@ -193,13 +204,11 @@ void RenderExecutor::ResolveRenderColorTarget(uint64_t submit_id, RenderCommandB
 	if (texture_tile &&
 	    (!TileGetTextureBlockLayout(transfer_format, rt.attrib3.tile_mode, volume,
 	                                texture_tile_layout) ||
-	     rt.pitch.pitch_div8_minus1 != 0 ||
 	     (rt.base.addr & (texture_tile_layout.block.block_size - 1u)) != 0 ||
 	     rt.info.fmask_compression_enable || rt.info.fmask_data_compression_disable ||
 	     rt.info.fmask_one_frag_mode || rt.info.cmask_fast_clear_enable ||
-	     rt.info.dcc_compression_enable || rt.info.cmask_is_linear != 0 ||
-	     rt.info.cmask_addr_type != 0 || rt.info.alt_tile_mode || rt.cmask.addr != 0 ||
-	     rt.fmask.addr != 0 || rt.dcc_addr.addr != 0 || rt.dcc.data_write_on_dcc_clear_to_reg)) {
+	     rt.info.dcc_compression_enable || rt.cmask.addr != 0 || rt.fmask.addr != 0 ||
+	     rt.dcc_addr.addr != 0 || rt.dcc.data_write_on_dcc_clear_to_reg)) {
 		EXIT("unsupported texture-tiled render target: addr=0x%016" PRIx64 " tile=%u"
 		     " dimension=%u depth=%u levels=%u layer=%u/%u samples=%u fragments=%u bpe=%u"
 		     " cmask=0x%016" PRIx64 " fmask=0x%016" PRIx64 " dcc=0x%016" PRIx64 "\n",
@@ -208,15 +217,14 @@ void RenderExecutor::ResolveRenderColorTarget(uint64_t submit_id, RenderCommandB
 		     rt.attrib.num_fragments, bytes_per_element, rt.cmask.addr, rt.fmask.addr,
 		     rt.dcc_addr.addr);
 	}
-	if ((standard64 || depth_tile) && (rt.attrib3.dimension != 1 || rt.attrib3.depth != 0 ||
-	                                   view.base_layer != 0 || view.image_layers != 1)) {
+	if ((standard64 || depth_tile) &&
+	    (rt.attrib3.dimension != 1 || rt.attrib3.depth != 0 ||
+	     (depth_tile && (view.base_layer != 0 || view.image_layers != 1)))) {
 		EXIT("unsupported 64KB texture-tiled render-target view: dimension=%u depth=%u"
 		     " layer=%u/%u\n",
 		     rt.attrib3.dimension, rt.attrib3.depth, view.base_layer, view.image_layers);
 	}
-	if (rt.pitch.pitch_div8_minus1 != 0) {
-		pitch = (rt.pitch.pitch_div8_minus1 + 1u) << 3u;
-	} else if (tile) {
+	if (tile) {
 		if (volume) {
 			pitch = TileGetTexturePitch(transfer_format, width, rt.attrib3.tile_mode);
 		} else if (texture_tile) {
@@ -282,12 +290,6 @@ void RenderExecutor::ResolveRenderColorTarget(uint64_t submit_id, RenderCommandB
 		mip_sizes[0]  = {static_cast<uint32_t>(size), 0, 0, 0, 0, 0};
 		mip_padded[0] = {pitch, height};
 	}
-	if (rt.slice.slice_div64_minus1 != 0 &&
-	    (static_cast<uint64_t>(rt.slice.slice_div64_minus1) + 1u) * 64u != size) {
-		EXIT("render-target slice span mismatch: encoded=0x%016" PRIx64 " derived=0x%016" PRIx64
-		     "\n",
-		     (static_cast<uint64_t>(rt.slice.slice_div64_minus1) + 1u) * 64u, size);
-	}
 	if (size == 0 || (!volume && size > UINT64_MAX / view.image_layers)) {
 		EXIT("render-target memory footprint is invalid\n");
 	}
@@ -297,7 +299,7 @@ void RenderExecutor::ResolveRenderColorTarget(uint64_t submit_id, RenderCommandB
 	if (backing_size == 0) {
 		EXIT("render-target backing is empty\n");
 	}
-	if (backing_size > TRACKER_ADDRESS_SIZE - rt.base.addr) {
+	if (!GuestRange {rt.base.addr, backing_size}.Valid()) {
 		EXIT("render-target backing range is invalid\n");
 	}
 
@@ -326,7 +328,7 @@ void RenderExecutor::ResolveRenderColorTarget(uint64_t submit_id, RenderCommandB
 	desc.info.data         = {rt.base.addr, backing_size};
 	desc.info.pixel_format = target_format.format;
 	desc.info.guest_format = transfer_format;
-	desc.info.type         = volume ? Prospero::ImageType::kColor3D : Prospero::ImageType::kColor2D;
+	desc.info.type         = image_type;
 	desc.info.extent       = {width, height, depth};
 	desc.info.resources    = {levels, volume ? 1u : view.image_layers};
 	desc.info.pitch        = pitch;
@@ -365,8 +367,13 @@ void RenderExecutor::ResolveRenderColorTarget(uint64_t submit_id, RenderCommandB
 		};
 	}
 	desc.view_info.format = target_format.format;
-	desc.view_info.type =
-	    view.layer_count == 1 ? vk::ImageViewType::e2D : vk::ImageViewType::e2DArray;
+	if (is_1d) {
+		desc.view_info.type =
+		    view.layer_count == 1 ? vk::ImageViewType::e1D : vk::ImageViewType::e1DArray;
+	} else {
+		desc.view_info.type =
+		    view.layer_count == 1 ? vk::ImageViewType::e2D : vk::ImageViewType::e2DArray;
+	}
 	desc.view_info.aspect      = vk::ImageAspectFlagBits::eColor;
 	desc.view_info.base_level  = rt.view.current_mip_level;
 	desc.view_info.level_count = 1;

src/graphics/host_gpu/renderer/colorRenderTarget.h

@@ -10,8 +10,6 @@
 
 namespace Libs::Graphics {
 
-class RenderCommandBuffer;
-
 enum class RenderColorType {
 	NoColorOutput,
 	RenderTexture,

src/graphics/host_gpu/renderer/commandScheduler.cpp

@@ -1,96 +1,97 @@
 #include "graphics/host_gpu/renderer/commandScheduler.h"
 
 #include "common/assert.h"
+#include "common/logging/log.h"
 #include "graphics/host_gpu/graphicContext.h"
 
 #include <algorithm>
+#include <cstdio>
+#include <optional>
 
 namespace Libs::Graphics {
 
 static thread_local CommandScheduler* g_deferred_callback_scheduler = nullptr;
 
-void CommandSlot::Reset() {
-	EXIT_IF(buffer == nullptr);
-	const auto result = buffer.reset(vk::CommandBufferResetFlagBits::eReleaseResources);
-	if (result != vk::Result::eSuccess) {
-		EXIT("failed to reset Vulkan command buffer: %s (%d)\n", VulkanToString(result).c_str(),
-		     static_cast<int>(result));
-	}
+namespace {
+
+void ReportVulkanFatal(const char* what, vk::Result result, uint64_t tick, uint32_t debug_op,
+                       uint64_t debug_submit, uint32_t arg0, uint32_t arg1, uint32_t arg2,
+                       uint32_t arg3, uint64_t arg4) {
+	LOGF("%s failed: %s (%d), tick=%" PRIu64 " debug_op=%u debug_submit=%" PRIu64
+	     " args=%u,%u,%u,%u,0x%016" PRIx64 "\n",
+	     what, VulkanToString(result).c_str(), static_cast<int>(result), tick, debug_op,
+	     debug_submit, arg0, arg1, arg2, arg3, arg4);
+	std::printf("%s failed: %s (%d), tick=%" PRIu64 " debug_op=%u debug_submit=%" PRIu64
+	            " args=%u,%u,%u,%u,0x%016" PRIx64 "\n",
+	            what, VulkanToString(result).c_str(), static_cast<int>(result), tick, debug_op,
+	            debug_submit, arg0, arg1, arg2, arg3, arg4);
+	std::fflush(stdout);
 }
 
-CommandScheduler::CommandPool::~CommandPool() {
-	Destroy();
-}
-
-void CommandScheduler::CommandPool::Create(GraphicContext& graphics) {
-	EXIT_IF(m_pool != nullptr || m_graphics != nullptr ||
-	        graphics.queue_family == static_cast<uint32_t>(-1));
-	m_graphics = &graphics;
+} // namespace
 
+CommandScheduler::CommandPool::CommandPool(GraphicContext& graphics, MasterSemaphore& master)
+    : m_graphics(graphics), m_master(master) {
+	EXIT_IF(graphics.queue_family == static_cast<uint32_t>(-1));
 	vk::CommandPoolCreateInfo create {};
 	create.sType            = vk::StructureType::eCommandPoolCreateInfo;
 	create.queueFamilyIndex = graphics.queue_family;
-	create.flags            = vk::CommandPoolCreateFlagBits::eResetCommandBuffer;
+	create.flags            = vk::CommandPoolCreateFlagBits::eTransient |
+	                          vk::CommandPoolCreateFlagBits::eResetCommandBuffer;
 	const auto result       = graphics.device.createCommandPool(&create, nullptr, &m_pool);
 	EXIT_NOT_IMPLEMENTED(result != vk::Result::eSuccess || m_pool == nullptr);
 }
 
-CommandSlot* CommandScheduler::CommandPool::CreateSlot() {
-	EXIT_IF(m_graphics == nullptr);
-	auto& graphics = *m_graphics;
+CommandScheduler::CommandPool::~CommandPool() {
+	m_graphics.device.destroyCommandPool(m_pool, nullptr);
+}
+
+size_t CommandScheduler::CommandPool::Grow() {
+	const auto first = m_ticks.size();
+	m_ticks.resize(first + GrowStep);
+	m_buffers.resize(first + GrowStep);
 
 	vk::CommandBufferAllocateInfo allocate {};
 	allocate.sType              = vk::StructureType::eCommandBufferAllocateInfo;
 	allocate.commandPool        = m_pool;
 	allocate.level              = vk::CommandBufferLevel::ePrimary;
-	allocate.commandBufferCount = 1;
-	vk::CommandBuffer buffer    = nullptr;
-	EXIT_IF(graphics.device.allocateCommandBuffers(&allocate, &buffer) != vk::Result::eSuccess);
-
-	vk::FenceCreateInfo fence_create {};
-	fence_create.sType = vk::StructureType::eFenceCreateInfo;
-	fence_create.flags = vk::FenceCreateFlagBits::eSignaled;
-	vk::Fence fence    = nullptr;
-	if (graphics.device.createFence(&fence_create, nullptr, &fence) != vk::Result::eSuccess) {
-		graphics.device.freeCommandBuffers(m_pool, 1, &buffer);
-		EXIT("failed to create command-buffer fence\n");
-	}
-
-	auto& slot      = m_slots.emplace_back();
-	slot.pool_mutex = &m_mutex;
-	slot.id         = static_cast<uint32_t>(m_slots.size() - 1);
-	slot.buffer     = buffer;
-	slot.fence      = fence;
-	return &slot;
+	allocate.commandBufferCount = static_cast<uint32_t>(GrowStep);
+	EXIT_IF(m_graphics.device.allocateCommandBuffers(&allocate, m_buffers.data() + first) !=
+	        vk::Result::eSuccess);
+	return first;
 }
 
-CommandSlot* CommandScheduler::CommandPool::Allocate(GraphicContext& graphics) {
-	Common::LockGuard lock(m_mutex);
-	if (m_pool == nullptr) {
-		Create(graphics);
-	}
-	EXIT_IF(m_graphics != &graphics);
-	auto  found = std::ranges::find_if(m_slots, [](const auto& slot) { return !slot.busy; });
-	auto* slot  = found != m_slots.end() ? &*found : CreateSlot();
-	slot->busy  = true;
-	slot->Reset();
-	return slot;
-}
+vk::CommandBuffer CommandScheduler::CommandPool::Commit() {
+	auto       gpu_tick = m_master.KnownGpuTick();
+	const auto search   = [this, &gpu_tick](size_t begin, size_t end) -> std::optional<size_t> {
+		for (size_t index = begin; index < end; ++index) {
+			if (gpu_tick >= m_ticks[index]) {
+				m_ticks[index] = m_master.CurrentTick();
+				return index;
+			}
+		}
+		return std::nullopt;
+	};
 
-void CommandScheduler::CommandPool::Destroy() {
-	Common::LockGuard lock(m_mutex);
-	if (m_pool == nullptr) {
-		return;
+	auto found = search(m_hint, m_ticks.size());
+	if (!found) {
+		m_master.Refresh();
+		gpu_tick = m_master.KnownGpuTick();
+		found    = search(m_hint, m_ticks.size());
 	}
-	EXIT_IF(std::ranges::any_of(m_slots, [](const auto& slot) { return slot.busy; }));
-	EXIT_IF(m_graphics == nullptr);
-	for (const auto& slot: m_slots) {
-		m_graphics->device.destroyFence(slot.fence, nullptr);
+	if (!found) {
+		found = search(0, m_hint);
 	}
-	m_graphics->device.destroyCommandPool(m_pool, nullptr);
-	m_slots.clear();
-	m_pool     = nullptr;
-	m_graphics = nullptr;
+	if (!found) {
+		found           = Grow();
+		m_ticks[*found] = m_master.CurrentTick();
+	}
+
+	m_hint = *found + 1;
+	if (m_hint == m_ticks.size()) {
+		m_hint = 0;
+	}
+	return m_buffers[*found];
 }
 
 bool CommandScheduler::InDeferredOperation() noexcept {
@@ -99,10 +100,8 @@ bool CommandScheduler::InDeferredOperation() noexcept {
 
 CommandScheduler::CommandScheduler(RenderContext& context, GraphicContext& graphics)
     : m_master(graphics), m_context(context), m_graphics(graphics),
-      m_priority_thread([this](std::stop_token stop) { PriorityOperationsThread(stop); }) {
-	m_buffers.reserve(CommandBufferGrowStep);
-	m_buffer_ticks.reserve(CommandBufferGrowStep);
-}
+      m_command_pool(graphics, m_master), m_command(*this),
+      m_priority_thread([this](std::stop_token stop) { PriorityOperationsThread(stop); }) {}
 
 CommandScheduler::~CommandScheduler() {
 	Shutdown();
@@ -128,9 +127,11 @@ void CommandScheduler::Shutdown() {
 		}
 		m_operation_state = OperationState::Draining;
 	}
-	if (Active() && m_recording) {
-		Finish();
+	if (!m_command.IsInvalid()) {
+		Submit();
 	}
+	m_master.Wait(CurrentTick() - 1);
+	PopPendingOperations();
 	DrainPriorityOperations();
 	m_priority_thread.request_stop();
 	m_operation_available.notify_all();
@@ -156,14 +157,10 @@ void CommandScheduler::Begin(HW::Context& registers, HW::UserConfig& user_config
 	m_user_config = &user_config;
 	m_shaders     = &shaders;
 
-	if (!Active()) {
-		m_current = static_cast<int>(GrowCommandBuffers());
-	}
-
-	BindCurrent();
-	if (!m_recording) {
-		Current().Begin();
-		m_recording = true;
+	if (m_command.IsInvalid()) {
+		BeginNext();
+	} else {
+		BindCurrent();
 	}
 }
 
@@ -172,7 +169,7 @@ void CommandScheduler::BeginRendering(const RenderState& state) {
 }
 
 void CommandScheduler::EndRendering() {
-	if (Active() && m_recording) {
+	if (Active() && !m_command.IsInvalid()) {
 		Current().EndRendering();
 	}
 }
@@ -183,55 +180,40 @@ void CommandScheduler::Flush() {
 }
 
 void CommandScheduler::Flush(SubmitInfo& submit) {
-	SubmitCurrent(submit);
+	Submit(submit);
 	BeginNext();
 }
 
-CommandBuffer& CommandScheduler::FlushAndGetSubmitted() {
-	SubmitInfo submit;
-	auto&      submitted = SubmitCurrent(submit);
+void CommandScheduler::FlushAndWait() {
+	const auto tick = Submit();
+	m_master.Wait(tick);
 	BeginNext();
-	return submitted;
 }
 
 void CommandScheduler::Finish() {
 	CheckActive();
-	const auto tick = CurrentTick();
-	if (m_recording) {
-		SubmitInfo submit;
-		SubmitCurrent(submit);
+	if (!m_command.IsInvalid()) {
+		Submit();
 	}
-	for (auto& buffer: m_buffers) {
-		buffer->WaitForFenceAndReset();
-	}
-	m_master.Wait(tick);
+	m_master.Wait(CurrentTick() - 1);
+	BeginNext();
 	PopPendingOperations();
-	BindCurrent();
-	Current().Begin();
-	m_recording = true;
-}
-
-void CommandScheduler::FinishCurrent() {
-	SubmitInfo submit;
-	auto&      submitted = SubmitCurrent(submit);
-	submitted.WaitForFenceAndReset();
-	m_master.Refresh();
-	PopPendingOperations();
-	submitted.Begin();
-	m_recording = true;
 }
 
 void CommandScheduler::Wait(uint64_t tick) {
-	CheckActive();
 	EXIT_IF(tick > CurrentTick());
-	if (tick >= CurrentTick()) {
+	if (tick == CurrentTick()) {
+		CheckActive();
 		// A stream-buffer wrap can wait while a draw is being prepared through a reference to
-		// Current(). Recycle the same command object so that reference remains valid.
-		FinishCurrent();
-		return;
+		// Current(). The wrapper stays stable while its pooled Vulkan buffer is retired. Deferred
+		// resources are released only at the next GPU operation boundary.
+		const auto submitted_tick = Submit();
+		EXIT_IF(submitted_tick != tick);
+		m_master.Wait(tick);
+		BeginNext();
+	} else {
+		m_master.Wait(tick);
 	}
-	m_master.Wait(tick);
-	PopPendingOperations();
 }
 
 void CommandScheduler::PopPendingOperations() {
@@ -243,17 +225,18 @@ void CommandScheduler::PopPendingOperations(bool refresh_gpu_tick) {
 		m_master.Refresh();
 	}
 	for (;;) {
-		Common::UniqueFunction<void> callback;
+		PendingOperation operation;
 		{
 			std::lock_guard lock(m_operation_mutex);
 			if (m_pending_operations.empty() ||
 			    !m_master.IsFree(m_pending_operations.front().tick)) {
 				return;
 			}
-			callback = std::move(m_pending_operations.front().callback);
+			operation = std::move(m_pending_operations.front());
 			m_pending_operations.pop();
 		}
-		RunOperation(std::move(callback));
+		WaitPriorityOperations(operation.tick);
+		RunOperation(std::move(operation.callback));
 	}
 }
 
@@ -359,86 +342,81 @@ bool CommandScheduler::IsFree(uint64_t tick) {
 	return m_master.IsFree(tick);
 }
 
-CommandSlot* CommandScheduler::AllocateCommandBuffer() {
-	return m_command_pool.Allocate(m_graphics);
-}
-
-uint64_t CommandScheduler::NextSubmitSequence() noexcept {
-	return m_submit_sequence.fetch_add(1, std::memory_order_relaxed) + 1;
-}
-
 void CommandScheduler::CheckActive() const {
-	EXIT_IF(!Active() || static_cast<size_t>(m_current) >= m_buffers.size());
+	EXIT_IF(!Active());
 }
 
-RenderCommandBuffer& CommandScheduler::Current() const {
+CommandBuffer& CommandScheduler::Current() {
 	CheckActive();
-	EXIT_IF(m_buffers[m_current] == nullptr);
-	return *m_buffers[m_current];
+	return m_command;
 }
 
-void CommandScheduler::BindCurrent() const {
+void CommandScheduler::BindCurrent() {
 	EXIT_IF(m_registers == nullptr || m_user_config == nullptr || m_shaders == nullptr);
-	Current().Bind(*m_registers, *m_user_config, *m_shaders);
+	m_command.Bind(*m_registers, *m_user_config, *m_shaders);
 }
 
-CommandBuffer& CommandScheduler::SubmitCurrent(SubmitInfo& submit) {
-	CheckActive();
-	EXIT_IF(!m_recording);
-	auto& submitted = Current();
-	submitted.End();
-	const auto signal_tick = m_master.NextTick();
-	submit.AddSignal(m_master.Handle(), signal_tick);
-	submitted.Execute(submit);
-	m_buffer_ticks[static_cast<size_t>(m_current)] = signal_tick;
-	m_recording                                    = false;
-	return submitted;
+CommandBuffer& CommandScheduler::BeginCommand() {
+	EXIT_IF(!m_command.IsInvalid());
+	m_command.m_buffer = m_command_pool.Commit();
+	m_command.Begin();
+	return m_command;
 }
 
-int CommandScheduler::FindReusableBuffer(uint64_t gpu_tick) const {
-	EXIT_IF(m_buffer_ticks.size() != m_buffers.size());
-	const auto buffer_count = m_buffers.size();
-	for (size_t offset = 1; offset <= buffer_count; ++offset) {
-		const auto candidate = (static_cast<size_t>(m_current) + offset) % buffer_count;
-		if (gpu_tick >= m_buffer_ticks[candidate]) {
-			return static_cast<int>(candidate);
-		}
+uint64_t CommandScheduler::Submit(SubmitInfo submit) {
+	EXIT_IF(m_command.IsInvalid());
+	EXIT_IF(submit.num_wait_semaphores > SubmitInfo::MaxSemaphores ||
+	        submit.num_signal_semaphores >= SubmitInfo::MaxSemaphores);
+
+	m_command.End();
+	const auto buffer   = m_command.m_buffer;
+	auto&      graphics = m_graphics;
+	EXIT_IF(graphics.queue == nullptr);
+
+	vk::Result result;
+	uint64_t   tick;
+	{
+		Common::LockGuard lock(graphics.queue_mutex);
+		tick = m_master.NextTick();
+		submit.AddSignal(m_master.Handle(), tick);
+
+		vk::TimelineSemaphoreSubmitInfo timeline_info {};
+		timeline_info.sType                     = vk::StructureType::eTimelineSemaphoreSubmitInfo;
+		timeline_info.waitSemaphoreValueCount   = submit.num_wait_semaphores;
+		timeline_info.pWaitSemaphoreValues      = submit.wait_ticks.data();
+		timeline_info.signalSemaphoreValueCount = submit.num_signal_semaphores;
+		timeline_info.pSignalSemaphoreValues    = submit.signal_ticks.data();
+
+		vk::SubmitInfo submit_info {};
+		submit_info.sType                = vk::StructureType::eSubmitInfo;
+		submit_info.pNext                = &timeline_info;
+		submit_info.waitSemaphoreCount   = submit.num_wait_semaphores;
+		submit_info.pWaitSemaphores      = submit.wait_semaphores.data();
+		submit_info.pWaitDstStageMask    = submit.wait_stages.data();
+		submit_info.commandBufferCount   = 1;
+		submit_info.pCommandBuffers      = &buffer;
+		submit_info.signalSemaphoreCount = submit.num_signal_semaphores;
+		submit_info.pSignalSemaphores    = submit.signal_semaphores.data();
+
+		result = graphics.queue.submit(1, &submit_info, nullptr);
 	}
-	return -1;
-}
 
-size_t CommandScheduler::GrowCommandBuffers() {
-	const auto first = m_buffers.size();
-	const auto end   = first + CommandBufferGrowStep;
-	m_buffers.reserve(end);
-	m_buffer_ticks.reserve(end);
-	for (size_t i = first; i < end; ++i) {
-		m_buffers.emplace_back(std::make_unique<RenderCommandBuffer>(*this));
-		m_buffer_ticks.push_back(0);
+	if (result != vk::Result::eSuccess) {
+		ReportVulkanFatal("vkQueueSubmit", result, tick, m_command.m_debug_op,
+		                  m_command.m_debug_submit_id, m_command.m_debug_arg0,
+		                  m_command.m_debug_arg1, m_command.m_debug_arg2, m_command.m_debug_arg3,
+		                  m_command.m_debug_arg4);
 	}
-	return first;
+	EXIT_NOT_IMPLEMENTED(result != vk::Result::eSuccess);
+
+	m_command.m_buffer = nullptr;
+	return tick;
 }
 
 void CommandScheduler::BeginNext() {
-	EXIT_IF(m_recording);
-
-	auto candidate = FindReusableBuffer(m_master.KnownGpuTick());
-	bool refreshed = false;
-	if (candidate < 0) {
-		m_master.Refresh();
-		refreshed = true;
-		candidate = FindReusableBuffer(m_master.KnownGpuTick());
-	}
-	if (candidate < 0) {
-		candidate = static_cast<int>(GrowCommandBuffers());
-	}
-
-	m_current = candidate;
-	Current().WaitForFenceAndReset();
-	PopPendingOperations(!refreshed);
+	EXIT_IF(!m_command.IsInvalid());
 	BindCurrent();
-	Current().Begin();
-	m_recording = true;
+	BeginCommand();
 }
 
 } // namespace Libs::Graphics

src/graphics/host_gpu/renderer/commandScheduler.h

@@ -6,11 +6,7 @@
 #include "graphics/host_gpu/renderer/masterSemaphore.h"
 #include "graphics/host_gpu/renderer/render.h"
 
-#include <array>
-#include <atomic>
 #include <condition_variable>
-#include <deque>
-#include <memory>
 #include <mutex>
 
 #include <queue>
@@ -20,16 +16,6 @@
 
 namespace Libs::Graphics {
 
-struct CommandSlot {
-	Common::Mutex*    pool_mutex = nullptr;
-	uint32_t          id         = 0;
-	vk::CommandBuffer buffer     = nullptr;
-	vk::Fence         fence      = nullptr;
-	bool              busy       = false;
-
-	void Reset();
-};
-
 class CommandScheduler {
 public:
 	CommandScheduler(RenderContext& context, GraphicContext& graphics);
@@ -41,9 +27,10 @@ public:
 	void           EndRendering();
 	void           Flush();
 	void           Flush(SubmitInfo& submit);
-	CommandBuffer& FlushAndGetSubmitted();
+	void           FlushAndWait();
 	void           Finish();
-	void           FinishCurrent();
+	CommandBuffer& BeginCommand();
+	uint64_t       Submit(SubmitInfo submit = {});
 	// Deferred callbacks can observe an externally owned drain, but cannot initiate shutdown:
 	// the priority runner cannot join itself.
 	void                      Shutdown();
@@ -55,34 +42,35 @@ public:
 	void                      DeferPriorityOperation(Common::UniqueFunction<void>&& operation);
 	[[nodiscard]] static bool InDeferredOperation() noexcept;
 
-	[[nodiscard]] bool            Active() const noexcept { return m_current >= 0; }
-	void                          CheckActive() const;
-	RenderCommandBuffer&          Current() const;
-	[[nodiscard]] uint64_t        CurrentTick() const noexcept { return m_master.CurrentTick(); }
-	[[nodiscard]] bool            IsFree(uint64_t tick);
-	[[nodiscard]] RenderContext&  Context() const noexcept { return m_context; }
-	[[nodiscard]] GraphicContext& Graphics() const noexcept { return m_graphics; }
+	[[nodiscard]] bool             Active() const noexcept { return m_registers != nullptr; }
+	void                           CheckActive() const;
+	CommandBuffer&                 Current();
+	[[nodiscard]] uint64_t         CurrentTick() const noexcept { return m_master.CurrentTick(); }
+	[[nodiscard]] bool             IsFree(uint64_t tick);
+	[[nodiscard]] MasterSemaphore& GetMasterSemaphore() noexcept { return m_master; }
+	[[nodiscard]] RenderContext&   Context() const noexcept { return m_context; }
+	[[nodiscard]] GraphicContext&  Graphics() const noexcept { return m_graphics; }
 
 private:
-	static constexpr size_t CommandBufferGrowStep = 4;
-
 	class CommandPool {
 	public:
-		CommandPool() = default;
+		CommandPool(GraphicContext& graphics, MasterSemaphore& master);
 		~CommandPool();
 		KYTY_CLASS_NO_COPY(CommandPool);
 
-		CommandSlot* Allocate(GraphicContext& graphics);
+		vk::CommandBuffer Commit();
 
 	private:
-		void         Create(GraphicContext& graphics);
-		CommandSlot* CreateSlot();
-		void         Destroy();
+		static constexpr size_t GrowStep = 4;
 
-		GraphicContext*         m_graphics = nullptr;
-		Common::Mutex           m_mutex;
-		vk::CommandPool         m_pool = nullptr;
-		std::deque<CommandSlot> m_slots;
+		size_t Grow();
+
+		GraphicContext&                m_graphics;
+		MasterSemaphore&               m_master;
+		vk::CommandPool                m_pool = nullptr;
+		std::vector<vk::CommandBuffer> m_buffers;
+		std::vector<uint64_t>          m_ticks;
+		size_t                         m_hint = 0;
 	};
 
 	enum class OperationState { Open, Draining, Closed };
@@ -92,39 +80,28 @@ private:
 		uint64_t                     tick = 0;
 	};
 
-	void                       BindCurrent() const;
-	CommandBuffer&             SubmitCurrent(SubmitInfo& submit);
-	void                       BeginNext();
-	void                       PopPendingOperations(bool refresh_gpu_tick);
-	[[nodiscard]] int          FindReusableBuffer(uint64_t gpu_tick) const;
-	[[nodiscard]] size_t       GrowCommandBuffers();
-	void                       PriorityOperationsThread(std::stop_token stop);
-	void                       RunOperation(Common::UniqueFunction<void>&& operation);
-	[[nodiscard]] CommandSlot* AllocateCommandBuffer();
-	[[nodiscard]] uint64_t     NextSubmitSequence() noexcept;
+	void BindCurrent();
+	void BeginNext();
+	void PopPendingOperations(bool refresh_gpu_tick);
+	void PriorityOperationsThread(std::stop_token stop);
+	void RunOperation(Common::UniqueFunction<void>&& operation);
 
-	MasterSemaphore                                   m_master;
-	RenderContext&                                    m_context;
-	GraphicContext&                                   m_graphics;
-	CommandPool                                       m_command_pool;
-	std::vector<std::unique_ptr<RenderCommandBuffer>> m_buffers;
-	std::vector<uint64_t>                             m_buffer_ticks;
-	std::queue<PendingOperation>                      m_pending_operations;
-	std::queue<PendingOperation>                      m_priority_operations;
-	std::mutex                                        m_operation_mutex;
-	std::condition_variable                           m_operation_available;
-	std::jthread                                      m_priority_thread;
-	bool                                              m_priority_active      = false;
-	uint64_t                                          m_priority_active_tick = 0;
-	OperationState                                    m_operation_state      = OperationState::Open;
-	int                                               m_current              = -1;
-	bool                                              m_recording            = false;
-	HW::Context*                                      m_registers            = nullptr;
-	HW::UserConfig*                                   m_user_config          = nullptr;
-	HW::Shader*                                       m_shaders              = nullptr;
-	std::atomic<uint64_t>                             m_submit_sequence      = 0;
-
-	friend class CommandBuffer;
+	MasterSemaphore              m_master;
+	RenderContext&               m_context;
+	GraphicContext&              m_graphics;
+	CommandPool                  m_command_pool;
+	CommandBuffer                m_command;
+	std::queue<PendingOperation> m_pending_operations;
+	std::queue<PendingOperation> m_priority_operations;
+	std::mutex                   m_operation_mutex;
+	std::condition_variable      m_operation_available;
+	std::jthread                 m_priority_thread;
+	bool                         m_priority_active      = false;
+	uint64_t                     m_priority_active_tick = 0;
+	OperationState               m_operation_state      = OperationState::Open;
+	HW::Context*                 m_registers            = nullptr;
+	HW::UserConfig*              m_user_config          = nullptr;
+	HW::Shader*                  m_shaders              = nullptr;
 };
 
 } // namespace Libs::Graphics

src/graphics/host_gpu/renderer/context.cpp

@@ -1,7 +1,5 @@
 #include "common/assert.h"
 #include "common/common.h"
-#include "common/emulatorConfig.h"
-#include "common/logging/log.h"
 #include "common/profiler.h"
 #include "common/threads.h"
 #include "graphics/host_gpu/graphicContext.h"
@@ -9,7 +7,6 @@
 #include "graphics/host_gpu/renderer/debug.h"
 #include "graphics/host_gpu/renderer/depthRenderTarget.h"
 #include "graphics/host_gpu/renderer/image/imageView.h"
-#include "graphics/host_gpu/renderer/pipeline/descriptorCache.h"
 #include "graphics/host_gpu/renderer/render.h"
 #include "graphics/host_gpu/renderer/renderContext.h"
 #include "graphics/host_gpu/vma.h"
@@ -17,118 +14,29 @@
 
 #include <algorithm>
 #include <bit>
-#include <cstdio>
 #include <cstring>
-#include <memory>
 namespace Libs::Graphics {
 
-namespace {
-
-void ReportVulkanFatal(const char* what, vk::Result result, uint32_t slot, uint64_t submit_seq,
-                       uint32_t debug_op, uint64_t debug_submit, uint32_t arg0, uint32_t arg1,
-                       uint32_t arg2, uint32_t arg3, uint64_t arg4) {
-	LOGF("%s failed: %s (%d), slot=%u submit_seq=%" PRIu64 " debug_op=%u debug_submit=%" PRIu64
-	     " args=%u,%u,%u,%u,0x%016" PRIx64 "\n",
-	     what, VulkanToString(result).c_str(), static_cast<int>(result), slot, submit_seq, debug_op,
-	     debug_submit, arg0, arg1, arg2, arg3, arg4);
-	std::printf("%s failed: %s (%d), slot=%u submit_seq=%" PRIu64 " debug_op=%u debug_submit=%" PRIu64
-	            " args=%u,%u,%u,%u,0x%016" PRIx64 "\n",
-	            what, VulkanToString(result).c_str(), static_cast<int>(result), slot, submit_seq,
-	            debug_op, debug_submit, arg0, arg1, arg2, arg3, arg4);
-	std::fflush(stdout);
-}
-
-} // namespace
-
-FenceResourceRetainer::~FenceResourceRetainer() {
-	if (!m_resources.empty()) {
-		EXIT("fence resource retainer destroyed before release\n");
-	}
-}
-
-void FenceResourceRetainer::Retain(std::shared_ptr<void> resource) {
-	if (resource == nullptr) {
-		EXIT("cannot retain a null fence resource\n");
-	}
-	if (std::ranges::none_of(m_resources, [&resource](const auto& retained) {
-		    return retained.get() == resource.get();
-	    })) {
-		m_resources.push_back(std::move(resource));
-	}
-}
-
-void FenceResourceRetainer::ReleaseAfterFence() noexcept {
-	m_resources.clear();
-}
-
 CommandBuffer::CommandBuffer(CommandScheduler& scheduler)
-    : m_context(scheduler.Context()), m_scheduler(scheduler), m_graphics(scheduler.Graphics()),
-      m_slot(scheduler.AllocateCommandBuffer()) {}
-
-CommandBuffer::~CommandBuffer() {
-	Release();
-}
+    : m_context(scheduler.Context()), m_graphics(scheduler.Graphics()) {}
 
 bool CommandBuffer::IsInvalid() const {
-	return m_slot == nullptr;
+	return m_buffer == nullptr;
 }
 
 vk::CommandBuffer CommandBuffer::Handle() const {
 	EXIT_IF(IsInvalid());
-
-	const auto handle = m_slot->buffer;
-	EXIT_IF(handle == nullptr);
-	return handle;
+	return m_buffer;
 }
 
-void CommandBuffer::Release() {
-	EXIT_IF(IsInvalid());
-
-	Common::LockGuard lock(*m_slot->pool_mutex);
-
-	WaitForFence();
-
-	m_slot->busy = false;
-	m_slot->Reset();
-	ReleaseResourcesAfterFence();
-	m_slot = nullptr;
-
-	EXIT_NOT_IMPLEMENTED(!IsInvalid());
-}
-
-void CommandBuffer::RetireBufferAfterFence(std::unique_ptr<VulkanBuffer> buffer) {
-	if (IsInvalid() || m_execute || buffer == nullptr || buffer->buffer == nullptr) {
-		EXIT("cannot retire a buffer on an invalid or submitted command buffer\n");
-	}
-	m_retired_buffers.push_back(std::move(buffer));
-}
-
-void CommandBuffer::RetainResourceUntilFence(std::shared_ptr<void> resource) {
-	if (IsInvalid() || m_execute) {
-		EXIT("cannot retain a resource on an invalid or submitted command buffer\n");
-	}
-	m_fence_resources.Retain(std::move(resource));
-}
-
-void CommandBuffer::RecycleDescriptorAfterFence(VulkanDescriptorSet& set) {
-	m_descriptor_sets_after_fence.push_back(&set);
-}
-
-void CommandBuffer::RecycleDescriptorsAfterFence() {
-	for (auto* set: m_descriptor_sets_after_fence) {
-		m_context.GetDescriptorCache().Recycle(*set);
-	}
-	m_descriptor_sets_after_fence.clear();
-}
-
-void CommandBuffer::Begin() const {
-	EXIT_IF(m_rendering);
+void CommandBuffer::Begin() {
+	EXIT_IF(m_rendering || IsInvalid());
 	auto buffer = Handle();
 
 	vk::CommandBufferBeginInfo begin_info {};
 	begin_info.sType            = vk::StructureType::eCommandBufferBeginInfo;
 	begin_info.pNext            = nullptr;
-	begin_info.flags            = {};
+	begin_info.flags            = vk::CommandBufferUsageFlagBits::eOneTimeSubmit;
 	begin_info.pInheritanceInfo = nullptr;
 
 	auto result = buffer.begin(&begin_info);
@@ -156,122 +64,6 @@ void CommandBuffer::SetDebugInfo(uint32_t op, uint64_t submit_id, uint32_t arg0,
 	m_debug_arg4      = arg4;
 }
 
-void CommandBuffer::Execute(const SubmitInfo& submit) {
-	EXIT_IF(IsInvalid());
-	EXIT_IF(m_execute);
-	EXIT_IF(submit.num_wait_semaphores > SubmitInfo::MaxSemaphores ||
-	        submit.num_signal_semaphores > SubmitInfo::MaxSemaphores);
-
-	auto buffer = Handle();
-	auto fence  = m_slot->fence;
-
-	vk::TimelineSemaphoreSubmitInfo timeline_info {};
-	timeline_info.sType                     = vk::StructureType::eTimelineSemaphoreSubmitInfo;
-	timeline_info.waitSemaphoreValueCount   = submit.num_wait_semaphores;
-	timeline_info.pWaitSemaphoreValues      = submit.wait_ticks.data();
-	timeline_info.signalSemaphoreValueCount = submit.num_signal_semaphores;
-	timeline_info.pSignalSemaphoreValues    = submit.signal_ticks.data();
-
-	vk::SubmitInfo submit_info {};
-	submit_info.sType                = vk::StructureType::eSubmitInfo;
-	submit_info.pNext                = &timeline_info;
-	submit_info.waitSemaphoreCount   = submit.num_wait_semaphores;
-	submit_info.pWaitSemaphores      = submit.wait_semaphores.data();
-	submit_info.pWaitDstStageMask    = submit.wait_stages.data();
-	submit_info.commandBufferCount   = 1;
-	submit_info.pCommandBuffers      = &buffer;
-	submit_info.signalSemaphoreCount = submit.num_signal_semaphores;
-	submit_info.pSignalSemaphores    = submit.signal_semaphores.data();
-
-	auto& graphics = m_graphics;
-	EXIT_IF(graphics.queue == nullptr);
-
-	auto result = graphics.device.resetFences(1, &fence);
-	if (result != vk::Result::eSuccess) {
-		ReportVulkanFatal("vkResetFences (before submit)", result, m_slot->id, m_submit_seq,
-		                  m_debug_op, m_debug_submit_id, m_debug_arg0, m_debug_arg1, m_debug_arg2,
-		                  m_debug_arg3, m_debug_arg4);
-	}
-	EXIT_NOT_IMPLEMENTED(result != vk::Result::eSuccess);
-
-	if (Config::GraphicsDebugDumpEnabled()) {
-		LOGF("vkQueueSubmit begin: slot=%u waits=%u signals=%u debug_op=%u debug_submit=%" PRIu64
-		     " args=%u,%u,%u,%u,0x%016" PRIx64 "\n",
-		     m_slot->id, submit.num_wait_semaphores, submit.num_signal_semaphores, m_debug_op,
-		     m_debug_submit_id, m_debug_arg0, m_debug_arg1, m_debug_arg2, m_debug_arg3,
-		     m_debug_arg4);
-	}
-
-	{
-		Common::LockGuard lock(graphics.queue_mutex);
-		m_submit_seq = m_scheduler.NextSubmitSequence();
-		result       = graphics.queue.submit(1, &submit_info, fence);
-	}
-
-	m_execute      = true;
-	m_fence_waited = false;
-
-	if (result != vk::Result::eSuccess) {
-		ReportVulkanFatal("vkQueueSubmit", result, m_slot->id, m_submit_seq, m_debug_op,
-		                  m_debug_submit_id, m_debug_arg0, m_debug_arg1, m_debug_arg2, m_debug_arg3,
-		                  m_debug_arg4);
-	}
-	EXIT_NOT_IMPLEMENTED(result != vk::Result::eSuccess);
-}
-
-void CommandBuffer::WaitForFence() {
-	FinalizeFence(false);
-}
-
-void CommandBuffer::WaitForFenceOnly() {
-	EXIT_IF(IsInvalid());
-	if (!m_execute || m_fence_waited) {
-		return;
-	}
-	auto device = m_graphics.device;
-	auto result = device.waitForFences(1, &m_slot->fence, VK_TRUE, UINT64_MAX);
-	if (result != vk::Result::eSuccess) {
-		ReportVulkanFatal("vkWaitForFences", result, m_slot->id, m_submit_seq, m_debug_op,
-		                  m_debug_submit_id, m_debug_arg0, m_debug_arg1, m_debug_arg2, m_debug_arg3,
-		                  m_debug_arg4);
-	}
-	EXIT_NOT_IMPLEMENTED(result != vk::Result::eSuccess);
-	m_fence_waited = true;
-}
-
-void CommandBuffer::WaitForFenceAndReset() {
-	FinalizeFence(true);
-}
-
-void CommandBuffer::FinalizeFence(bool reset_recording) {
-	const bool was_executed = m_execute;
-	WaitForFenceOnly();
-	if (was_executed) {
-		m_execute      = false;
-		m_fence_waited = false;
-		if (reset_recording) {
-			Common::LockGuard lock(*m_slot->pool_mutex);
-			m_slot->Reset();
-		}
-	}
-	if (was_executed) {
-		ReleaseResourcesAfterFence();
-	}
-	DeleteBuffersAfterFence();
-}
-
-void CommandBuffer::ReleaseResourcesAfterFence() {
-	RecycleDescriptorsAfterFence();
-	m_fence_resources.ReleaseAfterFence();
-}
-
-void CommandBuffer::DeleteBuffersAfterFence() {
-	for (const auto& buffer: m_retired_buffers) {
-		m_graphics.DeleteBuffer(*buffer);
-	}
-	m_retired_buffers.clear();
-}
-
 void CommandBuffer::BeginRendering(const RenderState& state) const {
 	EXIT_IF(state.width == 0 || state.height == 0 || state.num_layers == 0 ||
 	        state.num_color_attachments > RENDER_COLOR_ATTACHMENTS_MAX);

src/graphics/host_gpu/renderer/debug.cpp

@@ -27,7 +27,7 @@ static bool RenderTargetMaskHasMrt(uint32_t mask) {
 	return (mask & ~0x0fu) != 0;
 }
 
-static bool RenderTargetMaskHasBoundMrt(const RenderCommandBuffer& buffer) {
+static bool RenderTargetMaskHasBoundMrt(const CommandBuffer& buffer) {
 	const auto& hw   = buffer.GetRegisters();
 	const auto  mask = hw.GetRenderTargetMask();
 
@@ -45,7 +45,7 @@ static bool RenderTargetMaskHasBoundMrt(const RenderCommandBuffer& buffer) {
 	return bound_targets > 1;
 }
 
-uint32_t render_target_first_bound_slot(const RenderCommandBuffer& buffer) {
+uint32_t render_target_first_bound_slot(const CommandBuffer& buffer) {
 	const auto& hw   = buffer.GetRegisters();
 	const auto  mask = hw.GetRenderTargetMask();
 	for (uint32_t i = 0; i < 8; i++) {
@@ -84,17 +84,8 @@ void uc_print(const char* func, const HW::UserConfig& uc) {
 }
 
 void uc_check(const HW::UserConfig& uc) {
-	const auto& ge_cntl = uc.GetGeControl();
 	const auto& user_en = uc.GetGeUserVgprEn();
 
-	if (ge_cntl.primitive_group_size > 0x0040) {
-		LOGF("\t warning: unsupported GE_CNTL primitive_group_size = 0x%04" PRIx16 "\n",
-		     ge_cntl.primitive_group_size);
-	}
-	if (ge_cntl.vertex_group_size > 0x0040) {
-		LOGF("\t warning: unsupported GE_CNTL vertex_group_size = 0x%04" PRIx16 "\n",
-		     ge_cntl.vertex_group_size);
-	}
 	EXIT_NOT_IMPLEMENTED(user_en.vgpr1 != false);
 	EXIT_NOT_IMPLEMENTED(user_en.vgpr2 != false);
 	EXIT_NOT_IMPLEMENTED(user_en.vgpr3 != false);
@@ -106,17 +97,11 @@ void sh_print(const char* func, const HW::Shader& /*uc*/) {
 
 std::vector<std::string> rt_print(const char* func, const HW::RenderTarget& rt) {
 	std::vector<std::string> dst;
-	dst.reserve(53);
+	dst.reserve(40);
 
 	dst.push_back(fmt::format("{}\n", func));
 
 	dst.push_back(fmt::format("\t base.addr                       = 0x{:016x}\n", rt.base.addr));
-	dst.push_back(
-	    fmt::format("\t pitch.pitch_div8_minus1         = 0x{:08x}\n", rt.pitch.pitch_div8_minus1));
-	dst.push_back(fmt::format("\t pitch.fmask_pitch_div8_minus1   = 0x{:08x}\n",
-	                          rt.pitch.fmask_pitch_div8_minus1));
-	dst.push_back(fmt::format("\t slice.slice_div64_minus1        = 0x{:08x}\n",
-	                          rt.slice.slice_div64_minus1));
 	dst.push_back(fmt::format("\t view.base_array_slice_index     = 0x{:08x}\n",
 	                          rt.view.base_array_slice_index));
 	dst.push_back(fmt::format("\t view.last_array_slice_index     = 0x{:08x}\n",
@@ -137,12 +122,6 @@ std::vector<std::string> rt_print(const char* func, const HW::RenderTarget& rt)
 	                          rt.info.cmask_fast_clear_enable ? "true" : "false"));
 	dst.push_back(fmt::format("\t info.dcc_compression_enable     = {}\n",
 	                          rt.info.dcc_compression_enable ? "true" : "false"));
-	dst.push_back(
-	    fmt::format("\t info.cmask_is_linear            = 0x{:08x}\n", rt.info.cmask_is_linear));
-	dst.push_back(
-	    fmt::format("\t info.cmask_addr_type            = 0x{:08x}\n", rt.info.cmask_addr_type));
-	dst.push_back(fmt::format("\t info.alt_tile_mode              = {}\n",
-	                          rt.info.alt_tile_mode ? "true" : "false"));
 	dst.push_back(fmt::format("\t info.format                     = 0x{:08x}\n",
 	                          static_cast<uint32_t>(rt.info.format)));
 	dst.push_back(fmt::format("\t info.channel_type               = 0x{:08x}\n",
@@ -157,10 +136,6 @@ std::vector<std::string> rt_print(const char* func, const HW::RenderTarget& rt)
 	                          rt.info.round_mode ? "true" : "false"));
 	dst.push_back(fmt::format("\t attrib.force_dest_alpha_to_one  = {}\n",
 	                          rt.attrib.force_dest_alpha_to_one ? "true" : "false"));
-	dst.push_back(fmt::format("\t attrib.tile_mode                = 0x{:08x}\n",
-	                          static_cast<uint32_t>(rt.attrib.tile_mode)));
-	dst.push_back(fmt::format("\t attrib.fmask_tile_mode          = 0x{:08x}\n",
-	                          static_cast<uint32_t>(rt.attrib.fmask_tile_mode)));
 	dst.push_back(
 	    fmt::format("\t attrib.num_samples              = 0x{:08x}\n", rt.attrib.num_samples));
 	dst.push_back(
@@ -175,42 +150,32 @@ std::vector<std::string> rt_print(const char* func, const HW::RenderTarget& rt)
 	                          static_cast<uint32_t>(rt.attrib3.tile_mode)));
 	dst.push_back(
 	    fmt::format("\t attrib3.dimension               = 0x{:08x}\n", rt.attrib3.dimension));
-	dst.push_back(fmt::format("\t attrib3.cmask_pipe_aligned      = {}\n",
-	                          rt.attrib3.cmask_pipe_aligned ? "true" : "false"));
-	dst.push_back(fmt::format("\t attrib3.dcc_pipe_aligned        = {}\n",
-	                          rt.attrib3.dcc_pipe_aligned ? "true" : "false"));
+	dst.push_back(fmt::format("\t attrib3.metadata_pipe_aligned   = {}\n",
+	                          rt.attrib3.metadata_pipe_aligned ? "true" : "false"));
+	dst.push_back(fmt::format("\t attrib3.write_vrs_rate_hint_to_cmask = {}\n",
+	                          rt.attrib3.write_vrs_rate_hint_to_cmask ? "true" : "false"));
 	dst.push_back(fmt::format("\t dcc.max_uncompressed_block_size = 0x{:08x}\n",
 	                          rt.dcc.max_uncompressed_block_size));
 	dst.push_back(fmt::format("\t dcc.max_compressed_block_size   = 0x{:08x}\n",
 	                          rt.dcc.max_compressed_block_size));
-	dst.push_back(fmt::format("\t dcc.min_compressed_block_size   = 0x{:08x}\n",
-	                          rt.dcc.min_compressed_block_size));
 	dst.push_back(
 	    fmt::format("\t dcc.color_transform             = 0x{:08x}\n", rt.dcc.color_transform));
 	dst.push_back(fmt::format("\t dcc.overwrite_combiner_disable  = {}\n",
 	                          rt.dcc.overwrite_combiner_disable ? "true" : "false"));
-	dst.push_back(fmt::format("\t dcc.independent_64b_blocks      = {}\n",
-	                          rt.dcc.independent_64b_blocks ? "true" : "false"));
-	dst.push_back(fmt::format("\t dcc.independent_128b_blocks     = {}\n",
-	                          rt.dcc.independent_128b_blocks ? "true" : "false"));
+	dst.push_back(fmt::format("\t dcc.independent_block_size      = 0x{:02x}\n",
+	                          static_cast<uint8_t>(rt.dcc.independent_block_size)));
 	dst.push_back(fmt::format("\t data_write_on_dcc_clear_to_reg  = {}\n",
 	                          rt.dcc.data_write_on_dcc_clear_to_reg ? "true" : "false"));
 	dst.push_back(fmt::format("\t dcc.dcc_clear_key_enable        = {}\n",
 	                          rt.dcc.dcc_clear_key_enable ? "true" : "false"));
 	dst.push_back(fmt::format("\t cmask.addr                      = 0x{:016x}\n", rt.cmask.addr));
-	dst.push_back(fmt::format("\t cmask_slice.slice_minus1        = 0x{:08x}\n",
-	                          rt.cmask_slice.slice_minus1));
 	dst.push_back(fmt::format("\t fmask.addr                      = 0x{:016x}\n", rt.fmask.addr));
-	dst.push_back(fmt::format("\t fmask_slice.slice_minus1        = 0x{:08x}\n",
-	                          rt.fmask_slice.slice_minus1));
 	dst.push_back(
 	    fmt::format("\t clear_word0.word0               = 0x{:08x}\n", rt.clear_word0.word0));
 	dst.push_back(
 	    fmt::format("\t clear_word1.word1               = 0x{:08x}\n", rt.clear_word1.word1));
 	dst.push_back(
 	    fmt::format("\t dcc_addr.addr                   = 0x{:016x}\n", rt.dcc_addr.addr));
-	dst.push_back(fmt::format("\t size.width                      = 0x{:08x}\n", rt.size.width));
-	dst.push_back(fmt::format("\t size.height                     = 0x{:08x}\n", rt.size.height));
 
 	return dst;
 }
@@ -240,13 +205,8 @@ static bool RenderIsColorDimension(uint32_t dimension) {
 // NOLINTNEXTLINE(readability-function-cognitive-complexity)
 static void RtCheck(const HW::RenderTarget& rt) {
 	if (rt.base.addr != 0) {
-		// bool render_to_texture = (rt.attrib.tile_mode == 0x0d);
 		//  EXIT_NOT_IMPLEMENTED(rt.base_addr == 0);
 
-		EXIT_NOT_IMPLEMENTED(rt.pitch.pitch_div8_minus1 != 0);
-		EXIT_NOT_IMPLEMENTED(rt.pitch.fmask_pitch_div8_minus1 != 0);
-		EXIT_NOT_IMPLEMENTED(rt.slice.slice_div64_minus1 != 0);
-
 		EXIT_NOT_IMPLEMENTED(rt.view.base_array_slice_index > rt.view.last_array_slice_index);
 		if (rt.view.base_array_slice_index != 0x00000000 ||
 		    rt.view.last_array_slice_index != 0x00000000) {
@@ -266,12 +226,6 @@ static void RtCheck(const HW::RenderTarget& rt) {
 				logged = true;
 			}
 		}
-		const bool native_2x_one_frag = rt.info.fmask_compression_enable &&
-		                                !rt.info.fmask_data_compression_disable &&
-		                                rt.info.cmask_fast_clear_enable && rt.fmask.addr != 0 &&
-		                                rt.cmask.addr != 0 && rt.attrib.num_samples == 1 &&
-		                                rt.attrib.num_fragments == 1;
-		EXIT_NOT_IMPLEMENTED(rt.info.fmask_one_frag_mode && !native_2x_one_frag);
 		if (rt.info.fmask_compression_enable) {
 			EXIT_NOT_IMPLEMENTED(rt.attrib.num_samples == 0 && rt.attrib.num_fragments == 0);
 			static bool logged = false;
@@ -296,22 +250,6 @@ static void RtCheck(const HW::RenderTarget& rt) {
 				logged = true;
 			}
 		}
-		if (rt.info.cmask_is_linear != 0x00000000 || rt.info.cmask_addr_type != 0x00000000) {
-			static bool logged = false;
-			if (!logged) {
-				LOGF("RenderTarget: temporary: ignoring PS5 CMASK metadata fields "
-				     "is_linear=%u addr_type=%u\n",
-				     rt.info.cmask_is_linear, rt.info.cmask_addr_type);
-				logged = true;
-			}
-		}
-		if (rt.info.alt_tile_mode) {
-			static bool logged = false;
-			if (!logged) {
-				LOGF("RenderTarget: temporary: ignoring alternate tile mode flag\n");
-				logged = true;
-			}
-		}
 		if (rt.info.blend_bypass) {
 			static bool logged = false;
 			if (!logged) {
@@ -338,8 +276,6 @@ static void RtCheck(const HW::RenderTarget& rt) {
 				logged = true;
 			}
 		}
-		// EXIT_NOT_IMPLEMENTED(rt.tile_mode != 0x0000000a);
-		// EXIT_NOT_IMPLEMENTED(rt.fmask_tile_mode != 0x0000000a);
 		if (rt.attrib.num_samples != 0x00000000 || rt.attrib.num_fragments != 0x00000000) {
 			static bool logged = false;
 			if (!logged) {
@@ -374,34 +310,22 @@ static void RtCheck(const HW::RenderTarget& rt) {
 		if (!RenderIsColorDimension(rt.attrib3.dimension)) {
 			EXIT("unknown PS5 render-target dimension: 0x%08" PRIx32 "\n", rt.attrib3.dimension);
 		}
-		if (!rt.attrib3.cmask_pipe_aligned) {
+		if (!rt.attrib3.metadata_pipe_aligned) {
 			static bool logged = false;
 			if (!logged) {
-				LOGF("RenderTarget: temporary: accepting unaligned PS5 CMASK pipe flag\n");
-				logged = true;
-			}
-		}
-		if (!rt.attrib3.dcc_pipe_aligned) {
-			static bool logged = false;
-			if (!logged) {
-				LOGF("RenderTarget: temporary: accepting unaligned PS5 DCC pipe flag\n");
+				LOGF("RenderTarget: temporary: accepting unaligned PS5 metadata pipe flag\n");
 				logged = true;
 			}
 		}
+		EXIT_NOT_IMPLEMENTED(rt.attrib3.write_vrs_rate_hint_to_cmask);
 
 		// EXIT_NOT_IMPLEMENTED(rt.dcc_max_uncompressed_block_size != 0x00000002);
 		// EXIT_NOT_IMPLEMENTED(rt.dcc.max_compressed_block_size != 0x00000000);
-		EXIT_NOT_IMPLEMENTED(rt.dcc.min_compressed_block_size != 0x00000000);
 		// EXIT_NOT_IMPLEMENTED(rt.dcc.color_transform != 0x00000000);
 		EXIT_NOT_IMPLEMENTED(rt.dcc.overwrite_combiner_disable != false);
-		// EXIT_NOT_IMPLEMENTED(rt.dcc.force_independent_blocks != false);
-		// EXIT_NOT_IMPLEMENTED(rt.dcc.independent_128b_blocks != false);
 		// EXIT_NOT_IMPLEMENTED(rt.dcc.data_write_on_dcc_clear_to_reg != false);
 		EXIT_NOT_IMPLEMENTED(rt.dcc.dcc_clear_key_enable != false);
-		if (rt.cmask.addr != 0x0000000000000000 || rt.cmask_slice.slice_minus1 != 0x00000000 ||
-		    rt.fmask.addr != 0x0000000000000000 ||
-		    (rt.fmask_slice.slice_minus1 != 0x00000000 &&
-		     rt.fmask_slice.slice_minus1 != rt.slice.slice_div64_minus1) ||
+		if (rt.cmask.addr != 0x0000000000000000 || rt.fmask.addr != 0x0000000000000000 ||
 		    rt.dcc_addr.addr != 0x0000000000000000) {
 			static bool logged = false;
 			if (!logged) {
@@ -411,9 +335,6 @@ static void RtCheck(const HW::RenderTarget& rt) {
 				logged = true;
 			}
 		}
-
-		EXIT_NOT_IMPLEMENTED(rt.size.width != 0);
-		EXIT_NOT_IMPLEMENTED(rt.size.height != 0);
 	}
 }
 
@@ -433,37 +354,20 @@ static void ZPrint(const char* func, const HW::DepthRenderTarget& z) {
 	     "\t stencil_info.htile_stencil_disabled   = %s\n"
 	     "\t stencil_info.expclear_enabled         = %s\n"
 	     "\t stencil_info.partially_resident       = %s\n"
-	     "\t depth_info.addr5_swizzle_mask         = 0x%08" PRIx32 "\n"
-	     "\t depth_info.array_mode                 = 0x%08" PRIx32 "\n"
-	     "\t depth_info.pipe_config                = 0x%08" PRIx32 "\n"
-	     "\t depth_info.bank_width                 = 0x%08" PRIx32 "\n"
-	     "\t depth_info.bank_height                = 0x%08" PRIx32 "\n"
-	     "\t depth_info.macro_tile_aspect          = 0x%08" PRIx32 "\n"
-	     "\t depth_info.num_banks                  = 0x%08" PRIx32 "\n"
 	     "\t depth_view.slice_start                = 0x%08" PRIx32 "\n"
 	     "\t depth_view.slice_max                  = 0x%08" PRIx32 "\n"
 	     "\t depth_view.current_mip_level          = 0x%02" PRIx8 "\n"
 	     "\t depth_view.depth_write_disable        = %s\n"
 	     "\t depth_view.stencil_write_disable      = %s\n"
-	     "\t htile_surface.linear                  = 0x%08" PRIx32 "\n"
-	     "\t htile_surface.full_cache              = 0x%08" PRIx32 "\n"
-	     "\t htile_surface.htile_uses_preload_win  = 0x%08" PRIx32 "\n"
-	     "\t htile_surface.preload                 = 0x%08" PRIx32 "\n"
-	     "\t htile_surface.prefetch_width          = 0x%08" PRIx32 "\n"
-	     "\t htile_surface.prefetch_height         = 0x%08" PRIx32 "\n"
-	     "\t htile_surface.dst_outside_zero_to_one = 0x%08" PRIx32 "\n"
 	     "\t z_read_base_addr                      = 0x%016" PRIx64 "\n"
 	     "\t stencil_read_base_addr                = 0x%016" PRIx64 "\n"
 	     "\t z_write_base_addr                     = 0x%016" PRIx64 "\n"
 	     "\t stencil_write_base_addr               = 0x%016" PRIx64 "\n"
-	     "\t pitch_div8_minus1                     = 0x%08" PRIx32 "\n"
-	     "\t height_div8_minus1                    = 0x%08" PRIx32 "\n"
-	     "\t slice_div64_minus1                    = 0x%08" PRIx32 "\n"
 	     "\t htile_data_base_addr                  = 0x%016" PRIx64 "\n"
-	     "\t width                                 = 0x%08" PRIx32 "\n"
-	     "\t height                                = 0x%08" PRIx32 "\n"
+	     "\t shading_rate_encoding                 = 0x%02" PRIx8 "\n"
 	     "\t size.x_max                            = 0x%04" PRIx16 "\n"
-	     "\t size.y_max                            = 0x%04" PRIx16 "\n",
+	     "\t size.y_max                            = 0x%04" PRIx16 "\n"
+	     "\t size.valid                            = %s\n",
 	     static_cast<uint32_t>(z.z_info.format), z.z_info.num_samples,
 	     static_cast<uint32_t>(z.z_info.texture_compatibility),
 	     z.z_info.htile_acceleration ? "true" : "false",
@@ -474,30 +378,27 @@ static void ZPrint(const char* func, const HW::DepthRenderTarget& z) {
 	     static_cast<uint32_t>(z.stencil_info.texture_compatibility),
 	     z.stencil_info.htile_stencil_disabled ? "true" : "false",
 	     z.stencil_info.expclear_enabled ? "true" : "false",
-	     z.stencil_info.partially_resident ? "true" : "false", z.depth_info.addr5_swizzle_mask,
-	     z.depth_info.array_mode, z.depth_info.pipe_config, z.depth_info.bank_width,
-	     z.depth_info.bank_height, z.depth_info.macro_tile_aspect, z.depth_info.num_banks,
-	     z.depth_view.slice_start, z.depth_view.slice_max, z.depth_view.current_mip_level,
+	     z.stencil_info.partially_resident ? "true" : "false", z.depth_view.slice_start,
+	     z.depth_view.slice_max, z.depth_view.current_mip_level,
 	     z.depth_view.depth_write_disable ? "true" : "false",
-	     z.depth_view.stencil_write_disable ? "true" : "false", z.htile_surface.linear,
-	     z.htile_surface.full_cache, z.htile_surface.htile_uses_preload_win,
-	     z.htile_surface.preload, z.htile_surface.prefetch_width, z.htile_surface.prefetch_height,
-	     z.htile_surface.dst_outside_zero_to_one, z.z_read_base_addr, z.stencil_read_base_addr,
-	     z.z_write_base_addr, z.stencil_write_base_addr, z.pitch_div8_minus1, z.height_div8_minus1,
-	     z.slice_div64_minus1, z.htile_data_base_addr, z.width, z.height, z.size.x_max,
-	     z.size.y_max);
+	     z.depth_view.stencil_write_disable ? "true" : "false", z.z_read_base_addr,
+	     z.stencil_read_base_addr, z.z_write_base_addr, z.stencil_write_base_addr,
+	     z.htile_data_base_addr, z.shading_rate_encoding, z.size.x_max, z.size.y_max,
+	     z.size.valid ? "true" : "false");
 }
 
 // NOLINTNEXTLINE(readability-function-cognitive-complexity)
 static void ZCheck(const HW::DepthRenderTarget& z, const HW::DepthControl& dc,
                    const HW::RenderControl& rc) {
-	const bool depth_active =
-	    dc.z_enable || dc.z_write_enable || dc.depth_bounds_enable || rc.depth_clear_enable;
-	const bool stencil_active = dc.stencil_enable || rc.stencil_clear_enable;
+	const bool depth_active = dc.z_enable || dc.z_write_enable || dc.depth_bounds_enable ||
+	                          rc.depth_clear_enable || rc.copy_depth_to_color;
+	const bool stencil_active =
+	    dc.stencil_enable || rc.stencil_clear_enable || rc.copy_stencil_to_color;
 	if (!depth_active && !stencil_active) {
 		return;
 	}
 
+	EXIT_NOT_IMPLEMENTED(rc.copy_depth_to_color || rc.copy_stencil_to_color);
 	EXIT_NOT_IMPLEMENTED(!z.z_info.HasValidTextureCompatibility());
 	EXIT_NOT_IMPLEMENTED(!z.stencil_info.HasValidTextureCompatibility());
 	if (z.z_info.format == Prospero::DepthFormat::kInvalid) {
@@ -534,21 +435,7 @@ static void ZCheck(const HW::DepthRenderTarget& z, const HW::DepthControl& dc,
 
 	if (z.z_info.format != Prospero::DepthFormat::kInvalid ||
 	    z.stencil_info.format != Prospero::StencilFormat::kInvalid) {
-
-		EXIT_NOT_IMPLEMENTED(z.depth_info.addr5_swizzle_mask != 0x00000000);
-		EXIT_NOT_IMPLEMENTED(z.depth_info.array_mode != 0x00000000);
-		EXIT_NOT_IMPLEMENTED(z.depth_info.pipe_config != 0x00000000);
-		EXIT_NOT_IMPLEMENTED(z.depth_info.bank_width != 0x00000000);
-		EXIT_NOT_IMPLEMENTED(z.depth_info.bank_height != 0x00000000);
-		EXIT_NOT_IMPLEMENTED(z.depth_info.macro_tile_aspect != 0x00000000);
-		EXIT_NOT_IMPLEMENTED(z.depth_info.num_banks != 0x00000000);
-		EXIT_NOT_IMPLEMENTED(z.htile_surface.linear != 0x00000000);
-		EXIT_NOT_IMPLEMENTED(z.htile_surface.full_cache != 0x00000000);
-		EXIT_NOT_IMPLEMENTED(z.htile_surface.htile_uses_preload_win != 0x00000000);
-		EXIT_NOT_IMPLEMENTED(z.htile_surface.preload != 0x00000000);
-		EXIT_NOT_IMPLEMENTED(z.htile_surface.prefetch_width != 0x00000000);
-		EXIT_NOT_IMPLEMENTED(z.htile_surface.prefetch_height != 0x00000000);
-		EXIT_NOT_IMPLEMENTED(z.htile_surface.dst_outside_zero_to_one != 0x00000000);
+		EXIT_NOT_IMPLEMENTED(z.shading_rate_encoding != 0);
 
 		if (z.depth_view.current_mip_level != 0x00000000) {
 			static std::atomic<uint32_t> log_count {0};
@@ -570,23 +457,8 @@ static void ZCheck(const HW::DepthRenderTarget& z, const HW::DepthControl& dc,
 		EXIT_NOT_IMPLEMENTED(!z.depth_view.stencil_write_disable &&
 		                     z.stencil_read_base_addr != z.stencil_write_base_addr);
 		EXIT_NOT_IMPLEMENTED(!z.depth_view.depth_write_disable && z.z_write_base_addr == 0);
-		// EXIT_NOT_IMPLEMENTED(z.pitch_div8_minus1 != 0x000000ff);
-		// EXIT_NOT_IMPLEMENTED(z.height_div8_minus1 != 0x0000008f);
-		// EXIT_NOT_IMPLEMENTED(z.slice_div64_minus1 != 0x00008fff);
 		// EXIT_NOT_IMPLEMENTED(z.htile_data_base_addr == 0);
-		// EXIT_NOT_IMPLEMENTED(z.width != 0x00000780);
-		// EXIT_NOT_IMPLEMENTED(z.height != 0x00000438);
-
-		EXIT_NOT_IMPLEMENTED(z.width != 0);
-		EXIT_NOT_IMPLEMENTED(z.height != 0);
-		if (z.size.x_max == 0 || z.size.y_max == 0) {
-			static std::atomic<uint32_t> log_count {0};
-			if (log_count.fetch_add(1) < 16) {
-				LOGF("DepthTarget: temporary: inferring PS5 default depth extent x_max=0x%04" PRIx16
-				     ", y_max=0x%04" PRIx16 "\n",
-				     z.size.x_max, z.size.y_max);
-			}
-		}
+		EXIT_NOT_IMPLEMENTED(!z.size.valid);
 	}
 }
 
@@ -769,8 +641,6 @@ static void DPrint(const char* func, const HW::DepthControl& c, const HW::Stenci
 	     "\t backface_enable      = %s\n"
 	     "\t stencilfunc          = %" PRIu8 "\n"
 	     "\t stencilfunc_bf       = %" PRIu8 "\n"
-	     "\t color_writes_on_depth_fail_enable  = %s\n"
-	     "\t color_writes_on_depth_pass_disable = %s\n"
 	     "\t stencil_fail         = %" PRIu8 "\n"
 	     "\t stencil_zpass        = %" PRIu8 "\n"
 	     "\t stencil_zfail        = %" PRIu8 "\n"
@@ -787,12 +657,11 @@ static void DPrint(const char* func, const HW::DepthControl& c, const HW::Stenci
 	     "\t stencil_opval_bf     = %" PRIu8 "\n",
 	     c.stencil_enable ? "true" : "false", c.z_enable ? "true" : "false",
 	     c.z_write_enable ? "true" : "false", c.depth_bounds_enable ? "true" : "false", c.zfunc,
-	     c.backface_enable ? "true" : "false", c.stencilfunc, c.stencilfunc_bf,
-	     c.color_writes_on_depth_fail_enable ? "true" : "false",
-	     c.color_writes_on_depth_pass_disable ? "true" : "false", s.stencil_fail, s.stencil_zpass,
-	     s.stencil_zfail, s.stencil_fail_bf, s.stencil_zpass_bf, s.stencil_zfail_bf,
-	     sm.stencil_testval, sm.stencil_mask, sm.stencil_writemask, sm.stencil_opval,
-	     sm.stencil_testval_bf, sm.stencil_mask_bf, sm.stencil_writemask_bf, sm.stencil_opval_bf);
+	     c.backface_enable ? "true" : "false", c.stencilfunc, c.stencilfunc_bf, s.stencil_fail,
+	     s.stencil_zpass, s.stencil_zfail, s.stencil_fail_bf, s.stencil_zpass_bf,
+	     s.stencil_zfail_bf, sm.stencil_testval, sm.stencil_mask, sm.stencil_writemask,
+	     sm.stencil_opval, sm.stencil_testval_bf, sm.stencil_mask_bf, sm.stencil_writemask_bf,
+	     sm.stencil_opval_bf);
 }
 
 static void DCheck(const HW::DepthControl& c, const HW::StencilControl& s,
@@ -804,8 +673,6 @@ static void DCheck(const HW::DepthControl& c, const HW::StencilControl& s,
 	// Back-face stencil state is handled separately when enabled.
 	// EXIT_NOT_IMPLEMENTED(c.stencilfunc != 0);
 	// EXIT_NOT_IMPLEMENTED(c.stencilfunc_bf != 0);
-	EXIT_NOT_IMPLEMENTED(c.color_writes_on_depth_fail_enable != false);
-	EXIT_NOT_IMPLEMENTED(c.color_writes_on_depth_pass_disable != false);
 	// EXIT_NOT_IMPLEMENTED(s.stencil_fail != 0);
 	// EXIT_NOT_IMPLEMENTED(s.stencil_zpass != 0);
 	// EXIT_NOT_IMPLEMENTED(s.stencil_zfail != 0);
@@ -837,7 +704,12 @@ static void EqaaPrint(const char* func, const HW::EqaaControl& c) {
 	     c.static_anchor_associations ? "true" : "false");
 }
 
-static void EqaaCheck(const HW::EqaaControl& c) {
+static void EqaaCheck(const HW::EqaaControl& c, const HW::AaConfig& cf) {
+	// The EQAA controls only take effect while multisampling is on, so a single-sample
+	// target leaves them inert and has nothing to warn about.
+	if (cf.msaa_num_samples == 0) {
+		return;
+	}
 	if (c.max_anchor_samples != 0 || c.ps_iter_samples != 0 || c.mask_export_num_samples != 0 ||
 	    c.alpha_to_mask_num_samples != 0 || c.high_quality_intersections ||
 	    c.incoherent_eqaa_reads || c.interpolate_comp_z || c.static_anchor_associations) {
@@ -889,12 +761,10 @@ void LogDrawPhase(const char* draw_name, const char* phase) {
 	}
 }
 
-void LogPipelineTrace(const char* phase, uint32_t vs_hash0, uint32_t vs_crc32, uint32_t ps_hash0,
-                      uint32_t ps_crc32) {
+void LogPipelineTrace(const char* phase, uint64_t vertex_program_id, uint64_t pixel_program_id) {
 	if (graphics_debug_dump_enabled()) {
-		LOGF("PipelineTrace: %s VS=0x%08" PRIx32 "/0x%08" PRIx32 " PS=0x%08" PRIx32 "/0x%08" PRIx32
-		     "\n",
-		     phase, vs_hash0, vs_crc32, ps_hash0, ps_crc32);
+		LOGF("PipelineTrace: %s VS=%" PRIu64 " PS=%" PRIu64 "\n", phase, vertex_program_id,
+		     pixel_program_id);
 	}
 }
 
@@ -1158,7 +1028,7 @@ ScissorRect calc_final_scissor(const HW::ScreenViewport& vp, const HW::ScanModeC
 	return ScissorRectClamp(final, extent.width, extent.height);
 }
 
-void hw_check(const RenderCommandBuffer& buffer) {
+void hw_check(const CommandBuffer& buffer) {
 	const auto& hw      = buffer.GetRegisters();
 	const auto  rt_slot = render_target_first_bound_slot(buffer);
 	const auto& rt      = hw.GetRenderTarget(rt_slot);
@@ -1204,7 +1074,7 @@ void hw_check(const RenderCommandBuffer& buffer) {
 	log_phase("blend");
 	BcCheck(bc, bclr, cc);
 	log_phase("eqaa");
-	EqaaCheck(eqaa);
+	EqaaCheck(eqaa, ac);
 	log_phase("aa");
 	AaCheck(aa, ac);
 	log_phase("done");
@@ -1231,7 +1101,7 @@ void hw_check(const RenderCommandBuffer& buffer) {
 	// EXIT_NOT_IMPLEMENTED(hw.GetStencilClearValue() != 0);
 }
 
-void hw_print(const RenderCommandBuffer& buffer) {
+void hw_print(const CommandBuffer& buffer) {
 	const auto& hw      = buffer.GetRegisters();
 	const auto  rt_slot = render_target_first_bound_slot(buffer);
 	const auto& rt      = hw.GetRenderTarget(rt_slot);

src/graphics/host_gpu/renderer/debug.h

@@ -9,7 +9,7 @@
 
 namespace Libs::Graphics {
 
-class RenderCommandBuffer;
+class CommandBuffer;
 
 namespace HW {
 class Context;
@@ -28,15 +28,15 @@ struct ScissorRect {
 };
 
 uint32_t                 render_target_mask_slot(uint32_t mask, uint32_t slot);
-uint32_t                 render_target_first_bound_slot(const RenderCommandBuffer& buffer);
+uint32_t                 render_target_first_bound_slot(const CommandBuffer& buffer);
 bool                     graphics_debug_dump_enabled();
 void                     uc_print(const char* func, const HW::UserConfig& uc);
 void                     uc_check(const HW::UserConfig& uc);
 void                     sh_print(const char* func, const HW::Shader& uc);
 std::vector<std::string> rt_print(const char* func, const HW::RenderTarget& rt);
 bool                     RenderIsColorTileModeLinear(Prospero::TileMode tile_mode);
-void                     hw_print(const RenderCommandBuffer& buffer);
-void                     hw_check(const RenderCommandBuffer& buffer);
+void                     hw_print(const CommandBuffer& buffer);
+void                     hw_check(const CommandBuffer& buffer);
 void                     LogDrawPhase(const char* draw_name, const char* phase);
 ScissorRect calc_final_scissor(const HW::ScreenViewport& vp, const HW::ScanModeControl& smc,
                                vk::Extent2D extent);

src/graphics/host_gpu/renderer/depthRenderTarget.cpp

@@ -13,7 +13,6 @@
 #include "graphics/host_gpu/renderer/debug.h"
 #include "graphics/host_gpu/renderer/image/imageView.h"
 #include "graphics/host_gpu/renderer/image/textureCommon.h"
-#include "graphics/host_gpu/renderer/pipeline/descriptorCache.h"
 #include "graphics/host_gpu/renderer/render.h"
 #include "graphics/host_gpu/renderer/renderContext.h"
 #include "graphics/host_gpu/vulkanCommon.h"
@@ -71,8 +70,8 @@ static bool UsesStencilOpValue(uint8_t fail, uint8_t pass, uint8_t depth_fail) {
 	return fail == replace_op || pass == replace_op || depth_fail == replace_op;
 }
 
-[[nodiscard]] static vk::Format ResolveHostDepthAttachmentFormat(const RenderCommandBuffer& buffer,
-                                                                 const DepthFormatPolicy&   policy,
+[[nodiscard]] static vk::Format ResolveHostDepthAttachmentFormat(const CommandBuffer&     buffer,
+                                                                 const DepthFormatPolicy& policy,
                                                                  bool     has_stencil,
                                                                  uint32_t samples) {
 	auto&      graphics         = buffer.GetGraphics();
@@ -98,7 +97,7 @@ static bool UsesStencilOpValue(uint8_t fail, uint8_t pass, uint8_t depth_fail) {
 }
 
 // NOLINTNEXTLINE(readability-function-cognitive-complexity)
-void RenderExecutor::ResolveRenderDepthTarget(uint64_t submit_id, RenderCommandBuffer& buffer,
+void RenderExecutor::ResolveRenderDepthTarget(uint64_t submit_id, CommandBuffer& buffer,
                                               RenderDepthInfo& r) {
 	KYTY_PROFILER_FUNCTION();
 	(void)submit_id;
@@ -109,9 +108,10 @@ void RenderExecutor::ResolveRenderDepthTarget(uint64_t submit_id, RenderCommandB
 	const auto& sc          = hw.GetStencilControl();
 	const auto& sm          = hw.GetStencilMask();
 	const bool  has_stencil = z.stencil_info.format != Prospero::StencilFormat::kInvalid;
-	const bool  depth_active =
-	    dc.z_enable || dc.z_write_enable || dc.depth_bounds_enable || rc.depth_clear_enable;
-	const bool stencil_active = has_stencil && (dc.stencil_enable || rc.stencil_clear_enable);
+	const bool depth_active = dc.z_enable || dc.z_write_enable || dc.depth_bounds_enable ||
+	                          rc.depth_clear_enable || rc.copy_depth_to_color;
+	const bool stencil_active =
+	    has_stencil && (dc.stencil_enable || rc.stencil_clear_enable || rc.copy_stencil_to_color);
 	if (!depth_active && !stencil_active) {
 		return;
 	}
@@ -128,22 +128,14 @@ void RenderExecutor::ResolveRenderDepthTarget(uint64_t submit_id, RenderCommandB
 	    !z.stencil_info.expclear_enabled && !z.stencil_info.partially_resident &&
 	    z.depth_view.slice_start == 0 && z.depth_view.slice_max == 0 &&
 	    z.depth_view.current_mip_level == 0 && !z.depth_view.depth_write_disable &&
-	    !z.depth_view.stencil_write_disable && z.depth_info.addr5_swizzle_mask == 0 &&
-	    z.depth_info.array_mode == 0 && z.depth_info.pipe_config == 0 &&
-	    z.depth_info.bank_width == 0 && z.depth_info.bank_height == 0 &&
-	    z.depth_info.macro_tile_aspect == 0 && z.depth_info.num_banks == 0 &&
-	    z.htile_surface.linear == 0 && z.htile_surface.full_cache == 0 &&
-	    z.htile_surface.htile_uses_preload_win == 0 && z.htile_surface.preload == 0 &&
-	    z.htile_surface.prefetch_width == 0 && z.htile_surface.prefetch_height == 0 &&
-	    z.htile_surface.dst_outside_zero_to_one == 0 && z.z_read_base_addr == 0 &&
-	    z.z_write_base_addr == 0 && z.stencil_read_base_addr == 0 &&
-	    z.stencil_write_base_addr == 0 && z.htile_data_base_addr == 0 &&
+	    !z.depth_view.stencil_write_disable && z.z_read_base_addr == 0 && z.z_write_base_addr == 0 &&
+	    z.stencil_read_base_addr == 0 && z.stencil_write_base_addr == 0 &&
+	    z.htile_data_base_addr == 0 &&
 	    // DB_DEPTH_SIZE_XY is independent state and may remain programmed after the attachment
 	    // formats and addresses are unbound. A zero encoding is the valid 1x1 value, so its
 	    // presence alone must not manufacture a depth attachment.
-	    !z.z_info.htile_acceleration && !z.width_height_valid && !z.pitch_height_valid &&
-	    z.size.x_max == 0 && z.size.y_max == 0 && z.pitch_div8_minus1 == 0 &&
-	    z.height_div8_minus1 == 0 && z.slice_div64_minus1 == 0 && z.width == 0 && z.height == 0;
+	    !z.z_info.htile_acceleration && z.shading_rate_encoding == 0 && z.size.x_max == 0 &&
+	    z.size.y_max == 0;
 	if (attachment_unbound) {
 		static std::atomic_bool logged = false;
 		if (!logged.exchange(true, std::memory_order_relaxed)) {
@@ -166,17 +158,12 @@ void RenderExecutor::ResolveRenderDepthTarget(uint64_t submit_id, RenderCommandB
 			DepthFatal("invalid depth view: base=%u last=%u", z.depth_view.slice_start,
 			           z.depth_view.slice_max);
 	}
-	if (rc.copy_centroid || rc.copy_sample != 0 || z.z_info.expclear_enabled ||
+	if (rc.copy_depth_to_color || rc.copy_stencil_to_color || rc.copy_centroid ||
+	    rc.copy_sample != 0 || z.z_info.expclear_enabled ||
 	    z.stencil_info.expclear_enabled || z.z_info.partially_resident ||
 	    z.stencil_info.partially_resident || z.z_info.max_mip_level != 0 ||
-	    z.depth_view.current_mip_level != 0 || z.depth_info.addr5_swizzle_mask != 0 ||
-	    z.depth_info.array_mode != 0 || z.depth_info.pipe_config != 0 ||
-	    z.depth_info.bank_width != 0 || z.depth_info.bank_height != 0 ||
-	    z.depth_info.macro_tile_aspect != 0 || z.depth_info.num_banks != 0 ||
-	    z.htile_surface.linear != 0 || z.htile_surface.full_cache != 0 ||
-	    z.htile_surface.htile_uses_preload_win != 0 || z.htile_surface.preload != 0 ||
-	    z.htile_surface.prefetch_width != 0 || z.htile_surface.prefetch_height != 0 ||
-	    z.htile_surface.dst_outside_zero_to_one != 0 || z.z_read_base_addr == 0 ||
+	    z.depth_view.current_mip_level != 0 || z.shading_rate_encoding != 0 ||
+	    z.z_read_base_addr == 0 ||
 	    (!z.depth_view.depth_write_disable && z.z_write_base_addr != z.z_read_base_addr) ||
 	    (z.z_read_base_addr & 0xffffu) != 0 ||
 	    dc.zfunc > static_cast<uint8_t>(vk::CompareOp::eAlways)) {
@@ -190,10 +177,7 @@ void RenderExecutor::ResolveRenderDepthTarget(uint64_t submit_id, RenderCommandB
 		    (z.stencil_read_base_addr & 0xffffu) != 0) {
 			DepthFatal("unsupported stencil attachment state");
 		}
-	} else if (z.stencil_read_base_addr != 0 || z.stencil_write_base_addr != 0 ||
-	           !htile_stencil_compat ||
-	           z.stencil_info.texture_compatibility !=
-	               Prospero::TextureCompatibleStencil::kDisable) {
+	} else if (z.stencil_read_base_addr != 0 || z.stencil_write_base_addr != 0) {
 		DepthFatal("stencil state without an active stencil attachment");
 	}
 	if (has_htile) {
@@ -204,18 +188,13 @@ void RenderExecutor::ResolveRenderDepthTarget(uint64_t submit_id, RenderCommandB
 			DepthFatal("HTile clear tracking supports at most 32 slices");
 		}
 	}
-	const bool size_xy_valid = z.size.valid;
-	const bool wh_valid      = z.width_height_valid && z.width != 0 && z.height != 0;
-	if (!size_xy_valid && !wh_valid) {
+	if (!z.size.valid) {
 		DepthFatal("missing depth extent");
 	}
-	const uint32_t width  = size_xy_valid ? static_cast<uint32_t>(z.size.x_max) + 1u : z.width;
-	const uint32_t height = size_xy_valid ? static_cast<uint32_t>(z.size.y_max) + 1u : z.height;
-	if (width > 16384 || height > 16384 ||
-	    (size_xy_valid && wh_valid && (width != z.width || height != z.height)) ||
-	    (!z.pitch_height_valid &&
-	     (z.pitch_div8_minus1 != 0 || z.height_div8_minus1 != 0 || z.slice_div64_minus1 != 0))) {
-		DepthFatal("inconsistent depth extent or encoded layout");
+	const uint32_t width  = static_cast<uint32_t>(z.size.x_max) + 1u;
+	const uint32_t height = static_cast<uint32_t>(z.size.y_max) + 1u;
+	if (width > 16384 || height > 16384) {
+		DepthFatal("invalid depth extent");
 	}
 	const auto* policy = FindDepthFormatPolicy(z.z_info.format);
 	if (policy == nullptr) {
@@ -230,11 +209,6 @@ void RenderExecutor::ResolveRenderDepthTarget(uint64_t submit_id, RenderCommandB
 	const auto     guest_format = policy->guest_format;
 	const uint32_t bytes        = policy->bytes_per_element;
 	const auto     pitch        = TileGetDepthPitch(width, bytes, z.z_info.num_samples);
-	if (z.pitch_height_valid && ((static_cast<uint64_t>(z.pitch_div8_minus1) + 1u) * 8u != pitch ||
-	                             (static_cast<uint64_t>(z.height_div8_minus1) + 1u) * 8u !=
-	                                 ((static_cast<uint64_t>(height) + 7u) & ~7ull))) {
-		DepthFatal("encoded depth pitch or height mismatch");
-	}
 	TileSizeAlign depth_size {};
 	TileSizeAlign stencil_size {};
 	TileSizeAlign htile_size {};
@@ -245,14 +219,6 @@ void RenderExecutor::ResolveRenderDepthTarget(uint64_t submit_id, RenderCommandB
 	    (has_htile != (htile_size.align == 32768 && htile_size.size != 0))) {
 		DepthFatal("unsupported depth/stencil/HTile footprint");
 	}
-	if (z.pitch_height_valid &&
-	    (static_cast<uint64_t>(z.slice_div64_minus1) + 1u) * 64u != depth_size.size) {
-		DepthFatal("depth footprint mismatch: extent=%ux%u pitch=%u expected=0x%016" PRIx64
-		           " align=0x%016" PRIx64 " encoded_valid=%u encoded=0x%016" PRIx64,
-		           width, height, pitch, depth_size.size, depth_size.align,
-		           z.pitch_height_valid ? 1u : 0u,
-		           (static_cast<uint64_t>(z.slice_div64_minus1) + 1u) * 64u);
-	}
 	if (depth_size.size > UINT64_MAX / view.image_layers ||
 	    stencil_size.size > UINT64_MAX / view.image_layers ||
 	    htile_size.size > UINT64_MAX / view.image_layers) {
@@ -261,9 +227,9 @@ void RenderExecutor::ResolveRenderDepthTarget(uint64_t submit_id, RenderCommandB
 	const auto depth_backing_size   = depth_size.size * view.image_layers;
 	const auto stencil_backing_size = stencil_size.size * view.image_layers;
 	const auto htile_backing_size   = htile_size.size * view.image_layers;
-	if (depth_backing_size > TRACKER_ADDRESS_SIZE - z.z_read_base_addr ||
-	    (has_stencil && stencil_backing_size > TRACKER_ADDRESS_SIZE - z.stencil_read_base_addr) ||
-	    (has_htile && htile_backing_size > TRACKER_ADDRESS_SIZE - z.htile_data_base_addr)) {
+	if (!GuestRange {z.z_read_base_addr, depth_backing_size}.Valid() ||
+	    (has_stencil && !GuestRange {z.stencil_read_base_addr, stencil_backing_size}.Valid()) ||
+	    (has_htile && !GuestRange {z.htile_data_base_addr, htile_backing_size}.Valid())) {
 		DepthFatal("layered depth backing range is invalid");
 	}
 	r.htile                   = has_htile;
@@ -290,8 +256,8 @@ void RenderExecutor::ResolveRenderDepthTarget(uint64_t submit_id, RenderCommandB
 
 	r.stencil_clear_enable =
 	    has_stencil && rc.stencil_clear_enable && !z.depth_view.stencil_write_disable;
-	r.stencil_clear_value  = hw.GetStencilClearValue();
-	r.stencil_test_enable  = has_stencil && dc.stencil_enable;
+	r.stencil_clear_value = hw.GetStencilClearValue();
+	r.stencil_test_enable = has_stencil && dc.stencil_enable;
 	if (r.stencil_test_enable) {
 		const bool stencil_ops_disabled =
 		    rc.stencil_clear_enable || z.depth_view.stencil_write_disable;

src/graphics/host_gpu/renderer/depthRenderTarget.h

@@ -11,11 +11,11 @@
 
 namespace Libs::Graphics {
 
-class RenderCommandBuffer;
-
+// Without a stencil plane, Hi-Stencil fields are inactive. An active plane is compatible when
+// Hi-Stencil is disabled or HTile backing is present.
 inline constexpr bool depth_htile_stencil_acceleration_compatible(bool has_stencil, bool has_htile,
                                                                   bool htile_stencil_disabled) {
-	return htile_stencil_disabled || (has_stencil && has_htile);
+	return !has_stencil || htile_stencil_disabled || has_htile;
 }
 
 struct RenderDepthInfo {

src/graphics/host_gpu/renderer/image/image.cpp

@@ -73,10 +73,8 @@ namespace {
 	return usage;
 }
 
-void ValidateRange(GuestRange range, const char* name) {
-	if ((range.address == 0) != (range.size == 0) ||
-	    (range.address != 0 && (range.address >= TRACKER_ADDRESS_SIZE ||
-	                            range.size > TRACKER_ADDRESS_SIZE - range.address))) {
+void ValidateOptionalRange(GuestRange range, const char* name) {
+	if (!range.ValidOrEmpty()) {
 		EXIT("invalid %s image range: address=0x%016llx size=0x%016llx\n", name,
 		     static_cast<unsigned long long>(range.address),
 		     static_cast<unsigned long long>(range.size));
@@ -567,13 +565,13 @@ void Image::CopyMip(Image& source, uint32_t mip, uint32_t layer) {
 namespace ImageOps {
 
 void Validate(const ImageInfo& info) {
-	ValidateRange(info.data, "data");
-	ValidateRange(info.stencil, "stencil");
+	ValidateOptionalRange(info.data, "data");
+	ValidateOptionalRange(info.stencil, "stencil");
 
 	if (info.pixel_format == vk::Format::eUndefined) {
 		const bool metadata_empty =
-		    info.metadata.range.address == 0 && info.metadata.range.size == 0 &&
-		    info.metadata.kind == ImageMetadataKind::None && info.metadata.control == 0 &&
+		    info.metadata.range.Empty() && info.metadata.kind == ImageMetadataKind::None &&
+		    info.metadata.control == 0 &&
 		    info.metadata.compression == VideoOutCompression::Uncompressed &&
 		    !info.metadata.stencil_compressed;
 		if (info.data.Empty() || info.HasStencil() || !metadata_empty || info.extent.width == 0 ||
@@ -620,16 +618,14 @@ void Validate(const ImageInfo& info) {
 	}
 	switch (info.metadata.kind) {
 		case ImageMetadataKind::None:
-			if (info.metadata.range.address != 0 || info.metadata.range.size != 0 ||
-			    info.metadata.control != 0 ||
+			if (!info.metadata.range.Empty() || info.metadata.control != 0 ||
 			    info.metadata.compression != VideoOutCompression::Uncompressed ||
 			    info.metadata.stencil_compressed) {
 				EXIT("metadata-free image has metadata state\n");
 			}
 			break;
 		case ImageMetadataKind::Htile:
-			ValidateRange(info.metadata.range, "HTILE");
-			if (info.metadata.range.Empty() ||
+			if (!info.metadata.range.Valid() ||
 			    info.metadata.compression != VideoOutCompression::Uncompressed) {
 				EXIT("invalid HTILE metadata\n");
 			}
@@ -736,15 +732,10 @@ Image::~Image() {
 	if (m_graphics == nullptr) {
 		return;
 	}
-	{
-		std::lock_guard lock(views.mutex);
-		for (auto& cached: views.views) {
-			if (cached.view != nullptr) {
-				m_graphics->device.destroyImageView(cached.view, nullptr);
-				cached.view = nullptr;
-			}
+	for (const auto& cached: views) {
+		if (cached.view != nullptr) {
+			m_graphics->device.destroyImageView(cached.view, nullptr);
 		}
-		views.views.clear();
 	}
 	if (backing.image != nullptr) {
 		m_graphics->DeleteImage(backing);

src/graphics/host_gpu/renderer/image/image.h

@@ -2,12 +2,12 @@
 #define EMULATOR_SRC_GRAPHICS_HOST_GPU_RENDERER_IMAGE_H_
 
 #include "common/assert.h"
+#include "common/slotVector.h"
 #include "graphics/host_gpu/graphicContext.h"
 #include "graphics/host_gpu/renderer/image/imageInfo.h"
 
 #include <compare>
 #include <limits>
-#include <mutex>
 #include <optional>
 #include <span>
 #include <utility>
@@ -19,29 +19,13 @@ class Buffer;
 class CommandScheduler;
 struct ImageTestAccess;
 
-struct ImageId {
-	uint32_t index      = std::numeric_limits<uint32_t>::max();
-	uint32_t generation = 0;
-
-	[[nodiscard]] explicit operator bool() const noexcept {
-		return index != std::numeric_limits<uint32_t>::max();
-	}
-	auto operator<=>(const ImageId&) const = default;
-};
+using ImageId = Common::SlotId;
 
 struct CachedImageView {
 	ImageViewInfo info;
 	vk::ImageView view = nullptr;
 };
 
-struct ImageViewCache {
-	ImageViewCache() = default;
-	KYTY_CLASS_NO_COPY(ImageViewCache);
-
-	std::mutex                   mutex;
-	std::vector<CachedImageView> views;
-};
-
 struct ImageUsage {
 	bool texture       = false;
 	bool storage       = false;
@@ -157,7 +141,7 @@ public:
 
 	ImageInfo        info;
 	VulkanImage      backing;
-	ImageViewCache   views;
+	std::vector<CachedImageView> views;
 	ImageUsage       usage;
 	ImageBinding     binding;
 	bool             registered     = false;

src/graphics/host_gpu/renderer/image/imageInfo.h

@@ -15,18 +15,6 @@
 
 namespace Libs::Graphics {
 
-struct GuestRange {
-	uint64_t address = 0;
-	uint64_t size    = 0;
-
-	[[nodiscard]] constexpr bool Empty() const noexcept { return address == 0 || size == 0; }
-	[[nodiscard]] constexpr bool Valid() const noexcept {
-		return !Empty() && address < TRACKER_ADDRESS_SIZE && size <= TRACKER_ADDRESS_SIZE - address;
-	}
-	[[nodiscard]] constexpr uint64_t End() const noexcept { return address + size; }
-	auto                             operator<=>(const GuestRange&) const = default;
-};
-
 enum class VideoOutCompression : uint8_t { Uncompressed, Dcc256_256_0, Dcc256_64_64, Unsupported };
 
 enum class ImageMetadataKind : uint8_t { None, Htile, Dcc };
@@ -443,6 +431,10 @@ IsSupportedDisplayRenderTargetTileMode(Prospero::TileMode tile_mode) noexcept {
 		return encoded <= 0.04045f ? encoded / 12.92f : std::pow((encoded + 0.055f) / 1.055f, 2.4f);
 	};
 	switch (format) {
+		// A single-plane float target carries its clear as raw float bits, the same encoding the
+		// depth decoder below uses. Without this the clear is discarded and the target keeps stale
+		// contents.
+		case vk::Format::eR32Sfloat: next.float32[0] = std::bit_cast<float>(packed); break;
 		case vk::Format::eR32Uint: next.uint32[0] = packed; break;
 		case vk::Format::eR32Sint: next.int32[0] = static_cast<int32_t>(packed); break;
 		case vk::Format::eR8G8B8A8Srgb:

src/graphics/host_gpu/renderer/image/imageView.cpp

@@ -4,8 +4,6 @@
 #include "graphics/host_gpu/graphicContext.h"
 #include "graphics/host_gpu/renderer/image/image.h"
 
-#include <mutex>
-
 namespace Libs::Graphics {
 
 namespace {
@@ -357,8 +355,7 @@ vk::ImageView Image::FindView(const ImageViewInfo& view_info) {
 		     image.layers);
 	}
 
-	std::lock_guard lock(views.mutex);
-	for (const auto& cached: views.views) {
+	for (const auto& cached: views) {
 		if (cached.info == normalized) {
 			return cached.view;
 		}
@@ -394,7 +391,7 @@ vk::ImageView Image::FindView(const ImageViewInfo& view_info) {
 		     view_info.level_count, view_info.base_layer, view_info.layer_count,
 		     static_cast<vk::ImageUsageFlags::MaskType>(view_info.usage));
 	}
-	views.views.push_back({normalized, view});
+	views.push_back({normalized, view});
 	return view;
 }
 

src/graphics/host_gpu/renderer/pipeline/descriptorCache.cpp

@@ -1,318 +0,0 @@
-#include "graphics/host_gpu/renderer/pipeline/descriptorCache.h"
-
-#include "common/assert.h"
-#include "common/profiler.h"
-#include "graphics/shader/recompiler/ir/ShaderIR.h"
-
-#include <array>
-
-namespace Libs::Graphics {
-namespace {
-
-using BindingKind = ShaderRecompiler::IR::DescriptorBindingKind;
-
-vk::ShaderStageFlags StageFlags(DescriptorCache::Stage stage) {
-	switch (stage) {
-		case DescriptorCache::Stage::Vertex: return vk::ShaderStageFlagBits::eVertex;
-		case DescriptorCache::Stage::Pixel: return vk::ShaderStageFlagBits::eFragment;
-		case DescriptorCache::Stage::Compute: return vk::ShaderStageFlagBits::eCompute;
-		default: EXIT("unknown descriptor stage\n");
-	}
-}
-
-bool IsSampledImage(BindingKind kind) {
-	switch (kind) {
-		case BindingKind::Sampled1D:
-		case BindingKind::Sampled1DArray:
-		case BindingKind::Sampled2D:
-		case BindingKind::Sampled2DArray:
-		case BindingKind::Sampled2DMsaa:
-		case BindingKind::Sampled2DMsaaArray:
-		case BindingKind::Sampled3D:
-		case BindingKind::SampledUint1D:
-		case BindingKind::SampledUint1DArray:
-		case BindingKind::SampledUint2D:
-		case BindingKind::SampledUint2DArray:
-		case BindingKind::SampledUint2DMsaa:
-		case BindingKind::SampledUint2DMsaaArray:
-		case BindingKind::SampledUint3D: return true;
-		default: return false;
-	}
-}
-
-bool IsStorageImage(BindingKind kind) {
-	switch (kind) {
-		case BindingKind::Storage1D:
-		case BindingKind::Storage1DArray:
-		case BindingKind::Storage2D:
-		case BindingKind::Storage2DArray:
-		case BindingKind::Storage3D:
-		case BindingKind::StorageUint1D:
-		case BindingKind::StorageUint1DArray:
-		case BindingKind::StorageUint2D:
-		case BindingKind::StorageUint2DArray:
-		case BindingKind::StorageUint3D:
-		case BindingKind::StorageAtomic1D:
-		case BindingKind::StorageAtomic1DArray:
-		case BindingKind::StorageAtomic2D:
-		case BindingKind::StorageAtomic2DArray:
-		case BindingKind::StorageAtomic3D: return true;
-		default: return false;
-	}
-}
-
-vk::DescriptorType DescriptorType(BindingKind kind) {
-	if (kind == BindingKind::Samplers) {
-		return vk::DescriptorType::eSampler;
-	}
-	if (IsSampledImage(kind)) {
-		return vk::DescriptorType::eSampledImage;
-	}
-	if (IsStorageImage(kind)) {
-		return vk::DescriptorType::eStorageImage;
-	}
-	return vk::DescriptorType::eStorageBuffer;
-}
-
-uint32_t DescriptorCount(const ShaderRecompiler::IR::DescriptorBinding& binding) {
-	return binding.resources.empty() ? 1u : static_cast<uint32_t>(binding.resources.size());
-}
-
-std::vector<uint32_t> LayoutKey(DescriptorCache::Stage               stage,
-                                const ShaderRecompiler::IR::Program& program) {
-	std::vector<uint32_t> key;
-	key.reserve(1u + program.bindings.descriptors.size() * 4u);
-	key.push_back(static_cast<uint32_t>(stage));
-	for (const auto& binding: program.bindings.descriptors) {
-		key.push_back(static_cast<uint32_t>(binding.kind));
-		key.push_back(binding.binding);
-		key.push_back(DescriptorCount(binding));
-		key.push_back(static_cast<uint32_t>(DescriptorType(binding.kind)));
-	}
-	return key;
-}
-
-vk::DescriptorBufferInfo BufferInfo(const BufferView& view) {
-	EXIT_IF(view.buffer == nullptr);
-	return {view.buffer, view.offset, view.range};
-}
-
-} // namespace
-
-vk::DescriptorImageInfo DescriptorCache::MakeImageInfo(const TextureBinding& texture,
-                                                       uint32_t              element) {
-	const auto view =
-	    texture.mip_views.empty()
-	        ? (element == 0u ? texture.image_view : vk::ImageView {})
-	        : (element < texture.mip_views.size() ? texture.mip_views[element] : vk::ImageView {});
-	EXIT_IF(!texture.image_id || view == nullptr || texture.layout == vk::ImageLayout::eUndefined);
-	return {nullptr, view, texture.layout};
-}
-
-DescriptorCache::~DescriptorCache() {
-	for (auto& [layout, sets]: m_free_sets_by_layout) {
-		(void)layout;
-		for (auto* set: sets) {
-			delete set;
-		}
-	}
-	for (const auto& [key, layout]: m_descriptor_set_layouts) {
-		(void)key;
-		m_graphics.device.destroyDescriptorSetLayout(layout, nullptr);
-	}
-	for (const auto& pool: m_pools) {
-		m_graphics.device.destroyDescriptorPool(pool.pool, nullptr);
-	}
-}
-
-vk::DescriptorSetLayout
-DescriptorCache::GetDescriptorSetLayoutInternal(Stage                                stage,
-                                                const ShaderRecompiler::IR::Program& program) {
-	const auto key = LayoutKey(stage, program);
-	if (const auto found = m_descriptor_set_layouts.find(key);
-	    found != m_descriptor_set_layouts.end()) {
-		return found->second;
-	}
-
-	std::vector<vk::DescriptorSetLayoutBinding> bindings;
-	bindings.reserve(program.bindings.descriptors.size());
-	for (const auto& descriptor: program.bindings.descriptors) {
-		bindings.push_back({descriptor.binding, DescriptorType(descriptor.kind),
-		                    DescriptorCount(descriptor), StageFlags(stage), nullptr});
-	}
-	if (bindings.empty()) {
-		return nullptr;
-	}
-
-	vk::DescriptorSetLayoutCreateInfo info {};
-	info.sType                     = vk::StructureType::eDescriptorSetLayoutCreateInfo;
-	info.bindingCount              = static_cast<uint32_t>(bindings.size());
-	info.pBindings                 = bindings.data();
-	vk::DescriptorSetLayout layout = nullptr;
-	const auto result = m_graphics.device.createDescriptorSetLayout(&info, nullptr, &layout);
-	EXIT_NOT_IMPLEMENTED(result != vk::Result::eSuccess || layout == nullptr);
-	m_descriptor_set_layouts.emplace(key, layout);
-	return layout;
-}
-
-void DescriptorCache::CreatePool() {
-	KYTY_PROFILER_FUNCTION();
-	constexpr uint32_t           MaxSets = 512;
-	const vk::DescriptorPoolSize sizes[] = {
-	    {vk::DescriptorType::eStorageBuffer,
-	     MaxSets * (ShaderRecompiler::IR::ShaderInfo::MaxBuffers +
-	                ShaderRecompiler::IR::ShaderInfo::MaxAddresses + 3u)},
-	    {vk::DescriptorType::eSampledImage, MaxSets * ShaderRecompiler::IR::ShaderInfo::MaxImages},
-	    {vk::DescriptorType::eStorageImage, MaxSets * ShaderRecompiler::IR::ShaderInfo::MaxImages},
-	    {vk::DescriptorType::eSampler, MaxSets * ShaderRecompiler::IR::ShaderInfo::MaxSamplers},
-	};
-	vk::DescriptorPoolCreateInfo info {};
-	info.sType         = vk::StructureType::eDescriptorPoolCreateInfo;
-	info.poolSizeCount = static_cast<uint32_t>(std::size(sizes));
-	info.pPoolSizes    = sizes;
-	info.maxSets       = MaxSets;
-	const auto pool_id = static_cast<int>(m_pools.size());
-	auto&      pool    = m_pools.emplace_back();
-	const auto result  = m_graphics.device.createDescriptorPool(&info, nullptr, &pool.pool);
-	EXIT_NOT_IMPLEMENTED(result != vk::Result::eSuccess || pool.pool == nullptr);
-	pool.next_free_pool = m_first_free_pool;
-	m_first_free_pool   = pool_id;
-}
-
-VulkanDescriptorSet* DescriptorCache::Allocate(Stage                                stage,
-                                               const ShaderRecompiler::IR::Program& program) {
-	KYTY_PROFILER_FUNCTION();
-	Common::LockGuard lock(m_mutex);
-	const auto        layout = GetDescriptorSetLayoutInternal(stage, program);
-	EXIT_IF(layout == nullptr);
-	auto& free_sets = m_free_sets_by_layout[layout];
-	if (!free_sets.empty()) {
-		auto* result = free_sets.back();
-		free_sets.pop_back();
-		return result;
-	}
-
-	constexpr uint32_t Batch = 32;
-	for (int attempt = 0; attempt < 2; attempt++) {
-		for (int pool_id = m_first_free_pool; pool_id != -1;
-		     pool_id     = m_pools[pool_id].next_free_pool) {
-			auto&                                      pool = m_pools[pool_id];
-			std::array<vk::DescriptorSetLayout, Batch> layouts;
-			layouts.fill(layout);
-			std::array<vk::DescriptorSet, Batch> sets {};
-			vk::DescriptorSetAllocateInfo        info {};
-			info.sType              = vk::StructureType::eDescriptorSetAllocateInfo;
-			info.descriptorPool     = pool.pool;
-			info.descriptorSetCount = Batch;
-			info.pSetLayouts        = layouts.data();
-			if (m_graphics.device.allocateDescriptorSets(&info, sets.data()) ==
-			    vk::Result::eSuccess) {
-				free_sets.reserve(free_sets.size() + Batch - 1u);
-				for (uint32_t i = 0; i + 1u < Batch; i++) {
-					free_sets.push_back(new VulkanDescriptorSet {sets[i], layout, pool_id});
-				}
-				return new VulkanDescriptorSet {sets.back(), layout, pool_id};
-			}
-			m_first_free_pool = pool.next_free_pool;
-			break;
-		}
-		CreatePool();
-	}
-	return nullptr;
-}
-
-void DescriptorCache::Recycle(VulkanDescriptorSet& set) {
-	EXIT_IF(set.set == nullptr || set.layout == nullptr);
-	Common::LockGuard lock(m_mutex);
-	m_free_sets_by_layout[set.layout].push_back(&set);
-}
-
-VulkanDescriptorSet& DescriptorCache::GetDescriptor(Stage                                stage,
-                                                    const ShaderRecompiler::IR::Program& program,
-                                                    const NativeDescriptors&             data) {
-	KYTY_PROFILER_FUNCTION();
-	EXIT_IF(data.buffers.size() != program.info.buffers.size() ||
-	        data.images.size() != program.info.images.size() ||
-	        data.samplers.size() != program.info.samplers.size() ||
-	        data.addresses.size() != program.info.addresses.size());
-	auto* set = Allocate(stage, program);
-	EXIT_NOT_IMPLEMENTED(set == nullptr);
-
-	size_t descriptor_count = 0;
-	for (const auto& binding: program.bindings.descriptors) {
-		descriptor_count += DescriptorCount(binding);
-	}
-	std::vector<vk::DescriptorBufferInfo> buffer_infos;
-	std::vector<vk::DescriptorImageInfo>  image_infos;
-	std::vector<vk::WriteDescriptorSet>   writes;
-	buffer_infos.reserve(descriptor_count);
-	image_infos.reserve(descriptor_count);
-	writes.reserve(program.bindings.descriptors.size());
-	std::vector<uint32_t> image_occurrences(data.images.size());
-
-	for (const auto& binding: program.bindings.descriptors) {
-		vk::WriteDescriptorSet write {};
-		write.sType             = vk::StructureType::eWriteDescriptorSet;
-		write.dstSet            = set->set;
-		write.dstBinding        = binding.binding;
-		write.descriptorType    = DescriptorType(binding.kind);
-		write.descriptorCount   = DescriptorCount(binding);
-		const auto buffer_start = buffer_infos.size();
-		const auto image_start  = image_infos.size();
-		switch (binding.kind) {
-			case BindingKind::Buffers:
-				for (const auto resource: binding.resources) {
-					buffer_infos.push_back(BufferInfo(data.buffers.at(resource)));
-				}
-				break;
-			case BindingKind::AddressMemory:
-				for (const auto resource: binding.resources) {
-					buffer_infos.push_back(BufferInfo(data.addresses.at(resource)));
-				}
-				break;
-			case BindingKind::FlattenedSrt:
-				buffer_infos.push_back(BufferInfo(data.flattened_srt));
-				break;
-			case BindingKind::UserData: buffer_infos.push_back(BufferInfo(data.user_data)); break;
-			case BindingKind::Gds: buffer_infos.push_back(BufferInfo(data.gds)); break;
-			case BindingKind::Samplers:
-				for (const auto resource: binding.resources) {
-					const auto sampler = data.samplers.at(resource);
-					EXIT_IF(sampler == nullptr);
-					image_infos.push_back({sampler, nullptr, vk::ImageLayout::eUndefined});
-				}
-				break;
-			default: {
-				for (const auto resource: binding.resources) {
-					const auto& texture = data.images.at(resource);
-					image_infos.push_back(MakeImageInfo(texture, image_occurrences.at(resource)++));
-				}
-				break;
-			}
-		}
-		if (buffer_infos.size() != buffer_start) {
-			write.pBufferInfo = buffer_infos.data() + buffer_start;
-		}
-		if (image_infos.size() != image_start) {
-			write.pImageInfo = image_infos.data() + image_start;
-		}
-		writes.push_back(write);
-	}
-	for (uint32_t i = 0; i < data.images.size(); i++) {
-		const auto expected = data.images[i].mip_views.empty()
-		                          ? 1u
-		                          : static_cast<uint32_t>(data.images[i].mip_views.size());
-		EXIT_IF(image_occurrences[i] != expected);
-	}
-	m_graphics.device.updateDescriptorSets(static_cast<uint32_t>(writes.size()), writes.data(), 0,
-	                                       nullptr);
-	return *set;
-}
-
-vk::DescriptorSetLayout
-DescriptorCache::GetDescriptorSetLayout(Stage stage, const ShaderRecompiler::IR::Program& program) {
-	Common::LockGuard lock(m_mutex);
-	return GetDescriptorSetLayoutInternal(stage, program);
-}
-
-} // namespace Libs::Graphics

src/graphics/host_gpu/renderer/pipeline/descriptorCache.h

@@ -1,116 +0,0 @@
-#ifndef EMULATOR_SRC_GRAPHICS_HOST_GPU_RENDERER_DESCRIPTORCACHE_H_
-#define EMULATOR_SRC_GRAPHICS_HOST_GPU_RENDERER_DESCRIPTORCACHE_H_
-
-#include "common/abi.h"
-#include "common/assert.h"
-#include "common/common.h"
-#include "common/threads.h"
-#include "graphics/host_gpu/graphicContext.h"
-#include "graphics/host_gpu/renderer/cache/textureCache.h"
-#include "graphics/host_gpu/vulkanCommon.h"
-#include "graphics/shader/shaderBindings.h"
-
-#include <map>
-#include <memory>
-#include <span>
-#include <unordered_map>
-#include <vector>
-
-namespace Libs::Graphics {
-
-namespace ShaderRecompiler::IR {
-struct DescriptorValue;
-struct ImageResource;
-struct Program;
-struct ResourceSnapshot;
-} // namespace ShaderRecompiler::IR
-
-class CommandBuffer;
-struct DescriptorCacheTestAccess;
-struct ShaderStageRuntime;
-
-struct VulkanDescriptorSet {
-	vk::DescriptorSet       set     = nullptr;
-	vk::DescriptorSetLayout layout  = nullptr;
-	int                     pool_id = -1;
-};
-
-struct BufferView {
-	std::shared_ptr<void> owner;
-	vk::Buffer            buffer = nullptr;
-	vk::DeviceSize        offset = 0;
-	vk::DeviceSize        range  = VK_WHOLE_SIZE;
-};
-
-class DescriptorCache {
-public:
-	enum class Stage { Unknown, Vertex, Pixel, Compute };
-
-	struct TextureBinding {
-		ImageId                    image_id;
-		vk::ImageView              image_view = nullptr;
-		TextureCache::ImageDesc    desc;
-		vk::ImageLayout            layout = vk::ImageLayout::eUndefined;
-		std::vector<vk::ImageView> mip_views;
-	};
-
-	struct NativeDescriptors {
-		std::vector<BufferView>     buffers;
-		std::vector<TextureBinding> images;
-		std::vector<vk::Sampler>    samplers;
-		std::vector<BufferView>     addresses;
-		BufferView                  gds;
-		BufferView                  flattened_srt;
-		BufferView                  user_data;
-	};
-
-	struct PreparedBindings {
-		std::shared_ptr<const ShaderRecompiler::IR::Program>          program;
-		std::shared_ptr<const ShaderRecompiler::IR::ResourceSnapshot> snapshot;
-		NativeDescriptors                                             resources;
-		std::vector<uint32_t>                                         flattened_srt;
-		std::vector<uint32_t>                                         user_data;
-		vk::ShaderStageFlags                                          shader_stage;
-		Stage                                                         stage     = Stage::Unknown;
-		bool                                                          committed = false;
-	};
-
-	explicit DescriptorCache(GraphicContext& graphics): m_graphics(graphics) {
-		EXIT_NOT_IMPLEMENTED(!Common::Thread::IsMainThread());
-	}
-	~DescriptorCache();
-	KYTY_CLASS_NO_COPY(DescriptorCache);
-
-	vk::DescriptorSetLayout GetDescriptorSetLayout(Stage                                stage,
-	                                               const ShaderRecompiler::IR::Program& program);
-	void                    Recycle(VulkanDescriptorSet& set);
-	VulkanDescriptorSet&    GetDescriptor(Stage stage, const ShaderRecompiler::IR::Program& program,
-	                                      const NativeDescriptors& descriptors);
-
-private:
-	friend struct DescriptorCacheTestAccess;
-
-	struct Pool {
-		vk::DescriptorPool pool           = nullptr;
-		int                next_free_pool = -1;
-	};
-
-	static vk::DescriptorImageInfo MakeImageInfo(const TextureBinding& texture,
-	                                             uint32_t              element = 0);
-	void                           CreatePool();
-	VulkanDescriptorSet* Allocate(Stage stage, const ShaderRecompiler::IR::Program& program);
-	vk::DescriptorSetLayout
-	GetDescriptorSetLayoutInternal(Stage stage, const ShaderRecompiler::IR::Program& program);
-
-	GraphicContext&   m_graphics;
-	Common::Mutex     m_mutex;
-	std::vector<Pool> m_pools;
-	int               m_first_free_pool = -1;
-	std::unordered_map<vk::DescriptorSetLayout, std::vector<VulkanDescriptorSet*>>
-	                                                         m_free_sets_by_layout;
-	std::map<std::vector<uint32_t>, vk::DescriptorSetLayout> m_descriptor_set_layouts;
-};
-
-} // namespace Libs::Graphics
-
-#endif // EMULATOR_SRC_GRAPHICS_HOST_GPU_RENDERER_DESCRIPTORCACHE_H_

src/graphics/host_gpu/renderer/pipeline/descriptorHeap.cpp

@@ -0,0 +1,97 @@
+#include "graphics/host_gpu/renderer/pipeline/descriptorHeap.h"
+
+#include "common/assert.h"
+#include "common/profiler.h"
+#include "graphics/host_gpu/graphicContext.h"
+#include "graphics/host_gpu/renderer/masterSemaphore.h"
+
+namespace Libs::Graphics {
+namespace {
+
+constexpr uint32_t   DescriptorHeapCount = 1024;
+constexpr std::array DescriptorPoolSizes = {
+    vk::DescriptorPoolSize {vk::DescriptorType::eStorageBuffer, 8192},
+    vk::DescriptorPoolSize {vk::DescriptorType::eSampledImage, 8192},
+    vk::DescriptorPoolSize {vk::DescriptorType::eStorageImage, 1024},
+    vk::DescriptorPoolSize {vk::DescriptorType::eSampler, 1024},
+};
+
+} // namespace
+
+DescriptorHeap::DescriptorHeap(GraphicContext& graphics, MasterSemaphore& master_semaphore)
+    : m_graphics(graphics), m_master_semaphore(master_semaphore) {
+	CreatePool();
+}
+
+DescriptorHeap::~DescriptorHeap() {
+	m_graphics.device.destroyDescriptorPool(m_current_pool, nullptr);
+	for (const auto& [pool, tick]: m_pending_pools) {
+		m_master_semaphore.Wait(tick);
+		m_graphics.device.destroyDescriptorPool(pool, nullptr);
+	}
+}
+
+vk::DescriptorSet DescriptorHeap::Commit(vk::DescriptorSetLayout layout) {
+	KYTY_PROFILER_FUNCTION();
+	EXIT_IF(layout == nullptr);
+
+	auto& batch = m_sets[layout];
+	if (batch.size != 0) {
+		return batch.sets[--batch.size];
+	}
+	if (Allocate(layout, batch)) {
+		return batch.sets[--batch.size];
+	}
+
+	m_pending_pools.emplace_back(m_current_pool, m_master_semaphore.CurrentTick());
+	if (const auto& [pool, tick] = m_pending_pools.front(); m_master_semaphore.IsFree(tick)) {
+		m_current_pool = pool;
+		m_pending_pools.pop_front();
+		EXIT_IF(m_graphics.device.resetDescriptorPool(m_current_pool, {}) != vk::Result::eSuccess);
+	} else {
+		CreatePool();
+	}
+
+	m_sets.clear();
+	auto& fresh_batch = m_sets[layout];
+	EXIT_IF(!Allocate(layout, fresh_batch));
+	return fresh_batch.sets[--fresh_batch.size];
+}
+
+bool DescriptorHeap::Allocate(vk::DescriptorSetLayout layout, Batch& batch) {
+	std::array<vk::DescriptorSetLayout, DescriptorSetBatch> layouts;
+	layouts.fill(layout);
+
+	vk::DescriptorSetAllocateInfo allocate {};
+	allocate.sType          = vk::StructureType::eDescriptorSetAllocateInfo;
+	allocate.descriptorPool = m_current_pool;
+	allocate.pSetLayouts    = layouts.data();
+
+	for (;;) {
+		allocate.descriptorSetCount = batch.allocation;
+		const auto result = m_graphics.device.allocateDescriptorSets(&allocate, batch.sets.data());
+		if (result == vk::Result::eSuccess) {
+			batch.size = batch.allocation;
+			return true;
+		}
+		EXIT_IF(result != vk::Result::eErrorOutOfPoolMemory &&
+		        result != vk::Result::eErrorFragmentedPool);
+		if (batch.allocation == 1) {
+			return false;
+		}
+		batch.allocation /= 2;
+	}
+}
+
+void DescriptorHeap::CreatePool() {
+	vk::DescriptorPoolCreateInfo create {};
+	create.sType         = vk::StructureType::eDescriptorPoolCreateInfo;
+	create.flags         = {};
+	create.maxSets       = DescriptorHeapCount;
+	create.poolSizeCount = static_cast<uint32_t>(DescriptorPoolSizes.size());
+	create.pPoolSizes    = DescriptorPoolSizes.data();
+	EXIT_IF(m_graphics.device.createDescriptorPool(&create, nullptr, &m_current_pool) !=
+	        vk::Result::eSuccess);
+}
+
+} // namespace Libs::Graphics

src/graphics/host_gpu/renderer/pipeline/descriptorHeap.h

@@ -0,0 +1,45 @@
+#ifndef EMULATOR_SRC_GRAPHICS_HOST_GPU_RENDERER_DESCRIPTORHEAP_H_
+#define EMULATOR_SRC_GRAPHICS_HOST_GPU_RENDERER_DESCRIPTORHEAP_H_
+
+#include "common/common.h"
+#include "graphics/host_gpu/vulkanCommon.h"
+
+#include <array>
+#include <deque>
+#include <unordered_map>
+
+namespace Libs::Graphics {
+
+struct GraphicContext;
+class MasterSemaphore;
+
+class DescriptorHeap {
+public:
+	DescriptorHeap(GraphicContext& graphics, MasterSemaphore& master_semaphore);
+	~DescriptorHeap();
+	KYTY_CLASS_NO_COPY(DescriptorHeap);
+
+	[[nodiscard]] vk::DescriptorSet Commit(vk::DescriptorSetLayout layout);
+
+private:
+	static constexpr uint32_t DescriptorSetBatch = 32;
+
+	struct Batch {
+		std::array<vk::DescriptorSet, DescriptorSetBatch> sets {};
+		uint32_t                                          size       = 0;
+		uint32_t                                          allocation = DescriptorSetBatch;
+	};
+
+	[[nodiscard]] bool Allocate(vk::DescriptorSetLayout layout, Batch& batch);
+	void               CreatePool();
+
+	GraphicContext&                                     m_graphics;
+	MasterSemaphore&                                    m_master_semaphore;
+	vk::DescriptorPool                                  m_current_pool = nullptr;
+	std::deque<std::pair<vk::DescriptorPool, uint64_t>> m_pending_pools;
+	std::unordered_map<vk::DescriptorSetLayout, Batch>  m_sets;
+};
+
+} // namespace Libs::Graphics
+
+#endif // EMULATOR_SRC_GRAPHICS_HOST_GPU_RENDERER_DESCRIPTORHEAP_H_

src/graphics/host_gpu/renderer/pipeline/descriptors.cpp

@@ -17,7 +17,6 @@
 #include "graphics/host_gpu/renderer/debug.h"
 #include "graphics/host_gpu/renderer/image/imageView.h"
 #include "graphics/host_gpu/renderer/image/textureCommon.h"
-#include "graphics/host_gpu/renderer/pipeline/descriptorCache.h"
 #include "graphics/host_gpu/renderer/pipeline/shaderResourceBarrier.h"
 #include "graphics/host_gpu/renderer/render.h"
 #include "graphics/host_gpu/renderer/renderContext.h"
@@ -45,10 +44,99 @@
 
 namespace Libs::Graphics {
 
+namespace {
+
+using BindingKind = ShaderRecompiler::IR::DescriptorBindingKind;
+
+bool IsSampledImage(BindingKind kind) {
+	switch (kind) {
+		case BindingKind::Sampled1D:
+		case BindingKind::Sampled1DArray:
+		case BindingKind::Sampled2D:
+		case BindingKind::Sampled2DArray:
+		case BindingKind::Sampled2DMsaa:
+		case BindingKind::Sampled2DMsaaArray:
+		case BindingKind::Sampled3D:
+		case BindingKind::SampledUint1D:
+		case BindingKind::SampledUint1DArray:
+		case BindingKind::SampledUint2D:
+		case BindingKind::SampledUint2DArray:
+		case BindingKind::SampledUint2DMsaa:
+		case BindingKind::SampledUint2DMsaaArray:
+		case BindingKind::SampledUint3D: return true;
+		default: return false;
+	}
+}
+
+bool IsStorageImage(BindingKind kind) {
+	switch (kind) {
+		case BindingKind::Storage1D:
+		case BindingKind::Storage1DArray:
+		case BindingKind::Storage2D:
+		case BindingKind::Storage2DArray:
+		case BindingKind::Storage3D:
+		case BindingKind::StorageUint1D:
+		case BindingKind::StorageUint1DArray:
+		case BindingKind::StorageUint2D:
+		case BindingKind::StorageUint2DArray:
+		case BindingKind::StorageUint3D:
+		case BindingKind::StorageAtomic1D:
+		case BindingKind::StorageAtomic1DArray:
+		case BindingKind::StorageAtomic2D:
+		case BindingKind::StorageAtomic2DArray:
+		case BindingKind::StorageAtomic3D: return true;
+		default: return false;
+	}
+}
+
+} // namespace
+
+vk::DescriptorType NativeDescriptorType(BindingKind kind) {
+	if (kind == BindingKind::Samplers) {
+		return vk::DescriptorType::eSampler;
+	}
+	if (IsSampledImage(kind)) {
+		return vk::DescriptorType::eSampledImage;
+	}
+	if (IsStorageImage(kind)) {
+		return vk::DescriptorType::eStorageImage;
+	}
+	return vk::DescriptorType::eStorageBuffer;
+}
+
+uint32_t NativeDescriptorCount(const ShaderRecompiler::IR::DescriptorBinding& binding) {
+	return binding.resources.empty() ? 1u : static_cast<uint32_t>(binding.resources.size());
+}
+
+vk::DescriptorImageInfo MakeImageInfo(const TextureBinding& texture, uint32_t element) {
+	const auto view =
+	    texture.mip_views.empty()
+	        ? (element == 0u ? texture.image_view : vk::ImageView {})
+	        : (element < texture.mip_views.size() ? texture.mip_views[element] : vk::ImageView {});
+	EXIT_IF(!texture.image_id || view == nullptr || texture.layout == vk::ImageLayout::eUndefined);
+	return {nullptr, view, texture.layout};
+}
+
+static const char* ShaderStageResourceName(ShaderType stage) {
+	switch (stage) {
+		case ShaderType::Vertex: return "Vertex";
+		case ShaderType::Pixel: return "Pixel";
+		case ShaderType::Compute: return "Compute";
+		default: return "Unknown";
+	}
+}
+
+static vk::ShaderStageFlags NativeShaderStage(ShaderType stage) {
+	switch (stage) {
+		case ShaderType::Vertex: return vk::ShaderStageFlagBits::eVertex;
+		case ShaderType::Pixel: return vk::ShaderStageFlagBits::eFragment;
+		case ShaderType::Compute: return vk::ShaderStageFlagBits::eCompute;
+		default: EXIT("unknown native shader stage\n");
+	}
+}
+
 static void BindNullStorageBuffer(RenderContext& context, BufferView& dst) {
-	auto owner = context.GetBufferCache().ObtainNullBuffer();
-	dst.buffer = owner->Handle();
-	dst.owner  = std::move(owner);
+	dst.buffer = context.GetBufferCache().GetBuffer(NULL_BUFFER_ID).Handle();
 	dst.offset = 0;
 	dst.range  = 16;
 }
@@ -79,10 +167,11 @@ static bool IsMultisampledTexture(Prospero::ImageType type) {
 	       type == Prospero::ImageType::kColor2DMsaaArray;
 }
 
-static BufferView NativeStorageBuffer(RenderContext& context, CommandBuffer& command_buffer,
+static BufferView NativeStorageBuffer(RenderContext&                              context,
                                       const ShaderBufferResource&                 descriptor,
                                       const ShaderRecompiler::IR::BufferResource& resource,
-                                      uint32_t&                                   buffer_offset) {
+                                      ShaderType stage, uint32_t slot, uint32_t& buffer_offset,
+                                      BufferId id) {
 	BufferView result;
 	buffer_offset = 0;
 
@@ -104,60 +193,26 @@ static BufferView NativeStorageBuffer(RenderContext& context, CommandBuffer& com
 	    size > graphics.GetPhysicalDeviceProperties().limits.maxStorageBufferRange) {
 		EXIT("storage buffer range or device alignment is unsupported\n");
 	}
-	auto binding = context.GetBufferCache().ObtainBuffer(
-	    command_buffer, address, size, resource.written, resource.read, resource.formatted);
-	const auto aligned_offset = binding.offset - binding.offset % alignment;
-	const auto adjustment     = binding.offset - aligned_offset;
+	auto [buffer, offset] = context.GetBufferCache().ObtainBuffer(address, size, resource.written,
+	                                                              resource.formatted, id);
+	const auto aligned_offset = offset - offset % alignment;
+	const auto adjustment     = offset - aligned_offset;
 	const auto max_range      = graphics.GetPhysicalDeviceProperties().limits.maxStorageBufferRange;
 	if (adjustment % sizeof(uint32_t) != 0 || adjustment >= 256 || size > max_range - adjustment) {
 		EXIT("storage buffer offset adjustment is unsupported\n");
 	}
 	buffer_offset = static_cast<uint32_t>(adjustment);
-	result.owner  = std::move(binding.owner);
-	result.buffer = binding.buffer;
+	result.buffer = buffer->Handle();
 	result.offset = aligned_offset;
 	result.range  = static_cast<vk::DeviceSize>(size + adjustment);
-	return result;
-}
-
-static BufferView
-NativeAddressBuffer(RenderContext& context, CommandBuffer& command_buffer,
-                    const ShaderRecompiler::IR::AddressResource&           resource,
-                    const ShaderRecompiler::IR::ResourceSnapshot::Address& address,
-                    uint32_t&                                              address_offset) {
-	BufferView result;
-	address_offset = 0;
-	if (address.binding_base == 0) {
-		BindNullStorageBuffer(context, result);
-		return result;
+	if (resource.formatted && resource.written) {
+		context.GetTextureCache().InvalidateMemoryFromGPU(address, size);
 	}
-	if (resource.written) {
-		EXIT("writable address resources are unsupported\n");
-	}
-	const auto limit =
-	    resource.kind == ShaderRecompiler::IR::ResourceKind::Flat
-	        ? ShaderRecompiler::IR::FlatAddressWindowSize
-	        : static_cast<uint64_t>(
-	              context.GetGraphics().GetPhysicalDeviceProperties().limits.maxStorageBufferRange);
-	uint64_t   size   = 0;
-	const auto access = HostMemoryAccess::Mapped;
-	if (!HostMemoryQueryRange(address.binding_base, limit, access, size)) {
-		EXIT("address resource is not host-accessible: base=0x%016" PRIx64 "\n",
-		     address.binding_base);
-	}
-	const auto& graphics  = context.GetGraphics();
-	const auto  alignment = graphics.StorageMinAlignment();
-	auto binding =
-	    context.GetBufferCache().ObtainBuffer(command_buffer, address.binding_base, size);
-	const auto aligned_offset = binding.offset - binding.offset % alignment;
-	const auto adjustment     = binding.offset - aligned_offset;
-	const auto max_range      = graphics.GetPhysicalDeviceProperties().limits.maxStorageBufferRange;
-	const auto visible_size = std::min(size, static_cast<uint64_t>(max_range - adjustment));
-	address_offset = static_cast<uint32_t>(adjustment);
-	result.owner  = std::move(binding.owner);
-	result.buffer = binding.buffer;
-	result.offset = aligned_offset;
-	result.range  = static_cast<vk::DeviceSize>(visible_size + adjustment);
+	SetVulkanObjectNameF(
+	    graphics.device, result.buffer,
+	    "Kyty.{}.StorageBuffer[slot={} guest=0x{:016x} size=0x{:x} access={} formatted={}]",
+	    ShaderStageResourceName(stage), slot, address, size,
+	    resource.written ? (resource.read ? "ReadWrite" : "Write") : "Read", resource.formatted);
 	return result;
 }
 
@@ -458,10 +513,9 @@ void ValidateStorageTexture(const ShaderRecompiler::IR::ImageResource& resource,
 	const bool raw_sint_storage = format == Prospero::BufferFormat::k32SInt && uint_resource &&
 	                              resource.written && !resource.read && !resource.atomic;
 	const bool format_ok =
-	    raw_sint_storage ||
-	    (Prospero::IsSampledTextureFormat(format) &&
-	     uint_resource == Prospero::IsUintTextureFormat(format) &&
-	     (!resource.atomic || format == Prospero::BufferFormat::k32UInt));
+	    raw_sint_storage || (Prospero::IsSampledTextureFormat(format) &&
+	                         uint_resource == Prospero::IsUintTextureFormat(format) &&
+	                         (!resource.atomic || format == Prospero::BufferFormat::k32UInt));
 	if (resource_ok && descriptor_ok && encoding_ok && format_ok && size != 0) {
 		return;
 	}
@@ -568,8 +622,8 @@ static void PopulateTextureMipLayout(ImageInfo& info) {
 
 static ImageViewInfo TextureViewInfo(const ShaderRecompiler::IR::ImageResource& resource,
                                      const ShaderTextureResource& descriptor, vk::Format format,
-	                                 bool shader_conversion, bool storage,
-                                     uint32_t view_levels, uint32_t image_layers) {
+                                     bool shader_conversion, bool storage, uint32_t view_levels,
+                                     uint32_t image_layers) {
 	ImageViewInfo view {};
 	view.format      = format;
 	view.aspect      = vk::ImageAspectFlagBits::eColor;
@@ -624,9 +678,8 @@ static ImageViewInfo TextureViewInfo(const ShaderRecompiler::IR::ImageResource&
 	return view;
 }
 
-DescriptorCache::TextureBinding
-RenderExecutor::ResolveTexture(const ShaderRecompiler::IR::ImageResource&   resource,
-                               const ShaderRecompiler::IR::DescriptorValue& value) {
+TextureBinding RenderExecutor::ResolveTexture(const ShaderRecompiler::IR::ImageResource&   resource,
+                                              const ShaderRecompiler::IR::DescriptorValue& value) {
 	ShaderTextureResource descriptor;
 	CopyNativeDescriptor(value, descriptor.fields);
 	const bool storage = resource.written;
@@ -684,10 +737,9 @@ RenderExecutor::ResolveTexture(const ShaderRecompiler::IR::ImageResource&   reso
 	const bool shader_conversion =
 	    surface_format.conversion_format != Prospero::BufferFormat::kInvalid;
 	const bool sampled_numeric_class =
-	    storage ||
-	    (Prospero::IsSampledTextureFormat(format) &&
-	     (resource.kind == ShaderRecompiler::IR::ResourceKind::ImageUint) ==
-	         Prospero::IsUintTextureFormat(format));
+	    storage || (Prospero::IsSampledTextureFormat(format) &&
+	                (resource.kind == ShaderRecompiler::IR::ResourceKind::ImageUint) ==
+	                    Prospero::IsUintTextureFormat(format));
 	if (!storage &&
 	    (resource.kind == ShaderRecompiler::IR::ResourceKind::Image ||
 	     resource.kind == ShaderRecompiler::IR::ResourceKind::ImageUint) &&
@@ -752,7 +804,7 @@ RenderExecutor::ResolveTexture(const ShaderRecompiler::IR::ImageResource&   reso
 	                                 view_levels, desc.info.resources.layers);
 	desc.type = storage ? TextureCache::BindingType::Storage : TextureCache::BindingType::Texture;
 
-	auto id = texture_cache.FindImage(desc, shader_conversion);
+	auto       id                  = texture_cache.FindImage(desc, shader_conversion);
 	auto*      image               = &texture_cache.GetImage(id);
 	const bool stencil_association = static_cast<bool>(image->depth_id);
 	if (stencil_association) {
@@ -794,22 +846,19 @@ static vk::Sampler NativeSampler(RenderContext&                       context,
 	return context.GetSamplerCache().GetSampler(descriptor);
 }
 
-static BufferView NativeUpload(RenderContext& context, CommandBuffer& command_buffer,
-                               std::span<const uint32_t> data) {
+static BufferView NativeUpload(RenderContext& context, std::span<const uint32_t> data) {
 	EXIT_IF(data.empty());
-	EXIT_IF(command_buffer.IsInvalid() || command_buffer.IsExecute());
-	auto binding = context.GetBufferCache().UploadTransient(data.data(), data.size_bytes(), 256);
-	return {.owner  = std::move(binding.owner),
-	        .buffer = binding.buffer,
-	        .offset = binding.offset,
-	        .range  = data.size_bytes()};
+	auto& command_buffer = context.GetCommandScheduler().Current();
+	EXIT_IF(command_buffer.IsInvalid());
+	auto&      buffer = context.GetBufferCache().GetUtilityBuffer(MemoryUsage::Stream);
+	const auto offset = buffer.Copy(data.data(), data.size_bytes(), 256);
+	return {.buffer = buffer.Handle(), .offset = offset, .range = data.size_bytes()};
 }
 
 void RenderExecutor::TrackImageBinding(ImageId id) {
-	auto image = m_context.GetTextureCache().ResolveOwner(id);
-	EXIT_IF(image == nullptr);
-	if (std::ranges::find(m_bound_images, image) == m_bound_images.end()) {
-		m_bound_images.push_back(std::move(image));
+	EXIT_IF(m_context.GetTextureCache().m_slot_images.try_get(id) == nullptr);
+	if (std::ranges::find(m_bound_images, id) == m_bound_images.end()) {
+		m_bound_images.push_back(id);
 	}
 }
 
@@ -833,31 +882,28 @@ void RenderExecutor::BindRenderTarget(ImageId id) {
 }
 
 void RenderExecutor::ResetBindings() {
-	for (const auto& image: m_bound_images) {
-		image->binding = {};
+	for (const auto id: m_bound_images) {
+		if (auto* image = m_context.GetTextureCache().m_slot_images.try_get(id); image != nullptr) {
+			image->binding = {};
+		}
 	}
 	m_bound_images.clear();
 }
 
-DescriptorCache::PreparedBindings RenderExecutor::PrepareBindings(CommandBuffer&            buffer,
-                                                                  const ShaderStageRuntime& runtime,
-                                                                  vk::ShaderStageFlags shader_stage,
-                                                                  DescriptorCache::Stage stage) {
+PreparedBindings RenderExecutor::PrepareBindings(const ShaderStageRuntime& runtime) {
 	KYTY_PROFILER_FUNCTION();
 	EXIT_IF(!runtime);
 	const auto& program  = *runtime.program;
 	const auto& snapshot = *runtime.resources;
-	std::string error;
-	if (!ShaderRecompiler::IR::ValidateResourceSpecialization(program, snapshot, &error)) {
-		EXIT("invalid native shader runtime snapshot: %s\n", error.c_str());
+	if (!ShaderRecompiler::IR::ValidateResourceSpecialization(program, snapshot)) {
+		EXIT("invalid native shader runtime snapshot: hash=0x%016" PRIx64 " stage=%u\n",
+		     program.shader_hash, static_cast<unsigned>(program.stage));
 	}
 
-	DescriptorCache::PreparedBindings prepared;
-	prepared.program      = runtime.program;
-	prepared.snapshot     = runtime.resources;
-	prepared.shader_stage = shader_stage;
-	prepared.stage        = stage;
-	auto& descriptors     = prepared.resources;
+	PreparedBindings prepared;
+	prepared.program  = runtime.program.get();
+	prepared.snapshot = runtime.resources.get();
+	auto& descriptors = prepared.resources;
 	descriptors.buffers.reserve(program.info.buffers.size());
 	descriptors.images.reserve(program.info.images.size());
 	for (uint32_t i = 0; i < program.info.images.size(); i++) {
@@ -882,19 +928,46 @@ DescriptorCache::PreparedBindings RenderExecutor::PrepareBindings(CommandBuffer&
 	prepared.user_data.resize(program.bindings.ShaderDataDwords());
 	if (ShaderRecompiler::IR::FindBinding(
 	        program.bindings, ShaderRecompiler::IR::DescriptorBindingKind::Gds) != nullptr) {
-		descriptors.gds.buffer = m_context.GetBufferCache().GetGdsBuffer().Handle();
+		descriptors.gds.buffer = m_context.GetBufferCache().GetGdsBuffer()->Handle();
 	}
 	return prepared;
 }
 
-void RenderExecutor::RebindBuffers(CommandBuffer&                     buffer,
-                                   DescriptorCache::PreparedBindings& prepared) {
+void RenderExecutor::FindBuffers(PreparedBindings& prepared) {
+	KYTY_PROFILER_FUNCTION();
+	EXIT_IF(prepared.program == nullptr || prepared.snapshot == nullptr);
+	const auto& program  = *prepared.program;
+	const auto& snapshot = *prepared.snapshot;
+	auto&       cache    = m_context.GetBufferCache();
+
+	prepared.buffer_ids.clear();
+	prepared.buffer_ids.reserve(program.info.buffers.size());
+	for (uint32_t i = 0; i < program.info.buffers.size(); i++) {
+		ShaderBufferResource descriptor;
+		CopyNativeDescriptor(snapshot.buffers[i], descriptor.fields);
+		const auto address = descriptor.Base48();
+		const auto stride  = descriptor.Stride();
+		const auto records = descriptor.NumRecords();
+		EXIT_IF(stride != 0 && records > UINT64_MAX / stride);
+		const auto requested_size = stride != 0 ? static_cast<uint64_t>(stride) * records : records;
+		if (address == 0 || requested_size == 0) {
+			prepared.buffer_ids.emplace_back();
+			continue;
+		}
+		const auto size = Libs::LibKernel::Memory::ClampRangeSize(address, requested_size);
+		prepared.buffer_ids.push_back(cache.FindBuffer(address, size));
+	}
+
+}
+
+void RenderExecutor::RebindBuffers(PreparedBindings& prepared) {
 	KYTY_PROFILER_FUNCTION();
 	EXIT_IF(prepared.program == nullptr || prepared.snapshot == nullptr);
 	const auto& program   = *prepared.program;
 	const auto& snapshot  = *prepared.snapshot;
 	auto&       resources = prepared.resources;
 	const auto& layout    = program.bindings;
+	EXIT_IF(prepared.buffer_ids.size() != program.info.buffers.size());
 
 	resources.buffers.clear();
 	resources.buffers.reserve(program.info.buffers.size());
@@ -909,23 +982,21 @@ void RenderExecutor::RebindBuffers(CommandBuffer&                     buffer,
 		ShaderBufferResource descriptor;
 		CopyNativeDescriptor(snapshot.buffers[i], descriptor.fields);
 		uint32_t buffer_offset = 0;
-		resources.buffers.push_back(NativeStorageBuffer(m_context, buffer, descriptor,
-		                                                program.info.buffers[i], buffer_offset));
+		resources.buffers.push_back(NativeStorageBuffer(m_context, descriptor,
+		                                                program.info.buffers[i], program.stage, i,
+		                                                buffer_offset, prepared.buffer_ids[i]));
 		pack_memory_offset(i, buffer_offset);
 	}
-	resources.addresses.clear();
-	resources.addresses.reserve(program.info.addresses.size());
-	for (uint32_t i = 0; i < program.info.addresses.size(); i++) {
-		uint32_t address_offset = 0;
-		resources.addresses.push_back(NativeAddressBuffer(
-		    m_context, buffer, program.info.addresses[i], snapshot.addresses[i], address_offset));
-		pack_memory_offset(static_cast<uint32_t>(program.info.buffers.size()) + i,
-		                   address_offset);
+	if (!prepared.flattened_srt.empty()) {
+		resources.flattened_srt = NativeUpload(m_context, prepared.flattened_srt);
+	}
+	if (ShaderRecompiler::IR::FindBinding(
+	        program.bindings, ShaderRecompiler::IR::DescriptorBindingKind::UserData) != nullptr) {
+		resources.user_data = NativeUpload(m_context, prepared.user_data);
 	}
 }
 
-void RenderExecutor::RebindImages(CommandBuffer&                     buffer,
-                                  DescriptorCache::PreparedBindings& prepared) {
+void RenderExecutor::RebindImages(PreparedBindings& prepared) {
 	KYTY_PROFILER_FUNCTION();
 	EXIT_IF(prepared.program == nullptr || prepared.snapshot == nullptr);
 	const auto& program  = *prepared.program;
@@ -934,7 +1005,7 @@ void RenderExecutor::RebindImages(CommandBuffer&                     buffer,
 	EXIT_IF(images.size() != program.info.images.size());
 	auto& texture_cache = m_context.GetTextureCache();
 	for (uint32_t i = 0; i < program.info.images.size(); i++) {
-		const auto old_image = texture_cache.ResolveOwner(images[i].image_id);
+		const auto old_image = texture_cache.m_slot_images.try_get(images[i].image_id);
 		if (old_image == nullptr || (!old_image->registered && !old_image->info.data.Empty()) ||
 		    old_image->binding.needs_rebind) {
 			if (old_image != nullptr) {
@@ -944,6 +1015,8 @@ void RenderExecutor::RebindImages(CommandBuffer&                     buffer,
 			BindImage(images[i].image_id,
 			          images[i].desc.type == TextureCache::BindingType::Storage);
 		}
+	}
+	for (uint32_t i = 0; i < program.info.images.size(); i++) {
 		auto& binding = images[i];
 		binding.mip_views.clear();
 		const auto& resource = program.info.images[i];
@@ -973,121 +1046,221 @@ void RenderExecutor::RebindImages(CommandBuffer&                     buffer,
 }
 
 RenderExecutor::GraphicsBindings
-RenderExecutor::PrepareGraphicsBindings(CommandBuffer& buffer, const ShaderStageRuntime& vertex,
+RenderExecutor::PrepareGraphicsBindings(const ShaderStageRuntime& vertex,
                                         const ShaderStageRuntime& pixel, bool pixel_active) {
 	GraphicsBindings bindings {
-	    .vertex = PrepareBindings(buffer, vertex, vk::ShaderStageFlagBits::eVertex,
-	                              DescriptorCache::Stage::Vertex),
+	    .vertex = PrepareBindings(vertex),
 	};
-	RebindBuffers(buffer, bindings.vertex);
-	RebindImages(buffer, bindings.vertex);
 	if (pixel_active) {
-		bindings.pixel.emplace(PrepareBindings(buffer, pixel, vk::ShaderStageFlagBits::eFragment,
-		                                       DescriptorCache::Stage::Pixel));
-		RebindBuffers(buffer, *bindings.pixel);
-		RebindImages(buffer, *bindings.pixel);
+		bindings.pixel.emplace(PrepareBindings(pixel));
+	}
+	FindBuffers(bindings.vertex);
+	if (bindings.pixel) {
+		FindBuffers(*bindings.pixel);
+	}
+	if (bindings.vertex.program->info.uses_dma ||
+	    (bindings.pixel && bindings.pixel->program->info.uses_dma)) {
+		m_context.GetGpuResources().PrepareBda();
+	}
+	RebindBuffers(bindings.vertex);
+	if (bindings.pixel) {
+		RebindBuffers(*bindings.pixel);
+	}
+	RebindImages(bindings.vertex);
+	if (bindings.pixel) {
+		RebindImages(*bindings.pixel);
 	}
 	return bindings;
 }
 
-static void RetainBindings(CommandBuffer& buffer, DescriptorCache::NativeDescriptors& resources) {
-	if (resources.flattened_srt.owner != nullptr) {
-		buffer.RetainResourceUntilFence(std::move(resources.flattened_srt.owner));
-	}
-	if (resources.user_data.owner != nullptr) {
-		buffer.RetainResourceUntilFence(std::move(resources.user_data.owner));
-	}
-	for (auto& view: resources.buffers) {
-		if (view.owner != nullptr) {
-			buffer.RetainResourceUntilFence(std::move(view.owner));
-		}
-	}
-	for (auto& view: resources.addresses) {
-		if (view.owner != nullptr) {
-			buffer.RetainResourceUntilFence(std::move(view.owner));
-		}
-	}
-}
-
 void RenderExecutor::CommitBindings(CommandBuffer&                     buffer,
                                     vk::PipelineBindPoint              pipeline_bind_point,
-                                    vk::PipelineLayout                 layout,
-                                    DescriptorCache::PreparedBindings& prepared) {
+                                    const PipelineCache::Pipeline&     pipeline,
+                                    std::span<PreparedBindings* const> prepared_bindings) {
 	KYTY_PROFILER_FUNCTION();
-	EXIT_IF(prepared.program == nullptr || prepared.stage == DescriptorCache::Stage::Unknown ||
-	        prepared.committed);
-	const auto& program     = *prepared.program;
-	auto&       descriptors = prepared.resources;
-	if (!prepared.flattened_srt.empty()) {
-		descriptors.flattened_srt = NativeUpload(m_context, buffer, prepared.flattened_srt);
-	}
-	if (ShaderRecompiler::IR::FindBinding(
-	        program.bindings, ShaderRecompiler::IR::DescriptorBindingKind::UserData) != nullptr) {
-		descriptors.user_data = NativeUpload(m_context, buffer, prepared.user_data);
-	}
-	RetainBindings(buffer, descriptors);
-
-	auto       vk_buffer     = buffer.Handle();
-	const auto shader_stages = ShaderPipelineStages(prepared.shader_stage);
-	if (descriptors.gds.buffer != nullptr) {
-		buffer.EndRendering();
-		const auto barrier = MakeGdsDependency(descriptors.gds.buffer);
-		vk_buffer.pipelineBarrier(
-		    vk::PipelineStageFlagBits::eHost | vk::PipelineStageFlagBits::eTransfer |
-		        vk::PipelineStageFlagBits::eAllGraphics | vk::PipelineStageFlagBits::eComputeShader,
-		    shader_stages, vk::DependencyFlags {}, 0, nullptr, 1, &barrier, 0, nullptr);
-	}
-
-	for (uint32_t i = 0; i < program.info.images.size(); i++) {
-		auto&       image   = m_context.GetTextureCache().GetImage(descriptors.images[i].image_id);
-		auto&       binding = descriptors.images[i];
-		const auto& view    = binding.desc.view_info;
-		const ImageSubresourceRange range {view.base_level, view.level_count, view.base_layer,
-		                                   view.layer_count};
-		const bool storage = binding.desc.type == TextureCache::BindingType::Storage;
-		if (image.info.data.Empty()) {
-			image.Transit(vk::ImageLayout::eGeneral,
-			              storage
-			                  ? vk::AccessFlagBits2::eShaderRead | vk::AccessFlagBits2::eShaderWrite
-			                  : vk::AccessFlagBits2::eShaderRead,
-			              range, vk_buffer);
-		} else if ((image.binding.force_general || image.binding.is_target) &&
-		           !image.info.IsDepth()) {
-			const vk::AccessFlags2 storage_access = image.binding.shader_write
-			                                            ? vk::AccessFlagBits2::eShaderWrite
-			                                            : vk::AccessFlags2 {};
-			image.Transit(vk::ImageLayout::eGeneral,
-			              vk::AccessFlagBits2::eShaderRead | storage_access |
-			                  vk::AccessFlagBits2::eColorAttachmentRead |
-			                  vk::AccessFlagBits2::eColorAttachmentWrite,
-			              {}, vk_buffer);
-		} else if (storage) {
-			image.Transit(vk::ImageLayout::eGeneral,
-			              vk::AccessFlagBits2::eShaderRead | vk::AccessFlagBits2::eShaderWrite,
-			              range, vk_buffer);
-		} else {
-			image.Transit(image.info.IsDepth() ? vk::ImageLayout::eDepthStencilReadOnlyOptimal
-			                                   : vk::ImageLayout::eShaderReadOnlyOptimal,
-			              vk::AccessFlagBits2::eShaderRead, range, vk_buffer);
+	auto   vk_buffer        = buffer.Handle();
+	size_t descriptor_count = 0;
+	size_t write_count      = 0;
+	constexpr auto GraphicsStages =
+	    vk::ShaderStageFlagBits::eVertex | vk::ShaderStageFlagBits::eFragment;
+	const auto push_constant_stages = pipeline_bind_point == vk::PipelineBindPoint::eGraphics
+	                                      ? vk::ShaderStageFlags {GraphicsStages}
+	                                      : vk::ShaderStageFlags {vk::ShaderStageFlagBits::eCompute};
+	for (const auto* prepared: prepared_bindings) {
+		EXIT_IF(prepared == nullptr || prepared->program == nullptr ||
+		        prepared->snapshot == nullptr || prepared->committed);
+		write_count += prepared->program->bindings.descriptors.size();
+		for (const auto& binding: prepared->program->bindings.descriptors) {
+			descriptor_count += NativeDescriptorCount(binding);
 		}
-		binding.layout = image.backing.state.layout;
+		const auto shader_stage = NativeShaderStage(prepared->program->stage);
+		EXIT_IF((pipeline_bind_point == vk::PipelineBindPoint::eGraphics &&
+		         (shader_stage & GraphicsStages) == vk::ShaderStageFlags {}) ||
+		        (pipeline_bind_point == vk::PipelineBindPoint::eCompute &&
+		         shader_stage != vk::ShaderStageFlagBits::eCompute));
 	}
+	m_descriptor_buffers.clear();
+	m_descriptor_images.clear();
+	m_descriptor_writes.clear();
+	m_push_constants.fill(0);
+	m_descriptor_buffers.reserve(descriptor_count);
+	m_descriptor_images.reserve(descriptor_count);
+	m_descriptor_writes.reserve(write_count);
 
-	if (!program.bindings.descriptors.empty()) {
-		auto& set =
-		    m_context.GetDescriptorCache().GetDescriptor(prepared.stage, program, descriptors);
-		vk_buffer.bindDescriptorSets(pipeline_bind_point, layout, program.bindings.descriptor_set,
-		                             1, &set.set, 0, nullptr);
-		buffer.RecycleDescriptorAfterFence(set);
+	for (auto* prepared: prepared_bindings) {
+		const auto& program       = *prepared->program;
+		auto&       descriptors   = prepared->resources;
+		const auto  shader_stage  = NativeShaderStage(program.stage);
+		const auto  shader_stages = ShaderPipelineStages(shader_stage);
+		if (descriptors.gds.buffer != nullptr) {
+			buffer.EndRendering();
+			const auto barrier = MakeGdsDependency(descriptors.gds.buffer);
+			vk_buffer.pipelineBarrier(
+			    vk::PipelineStageFlagBits::eHost | vk::PipelineStageFlagBits::eTransfer |
+			        vk::PipelineStageFlagBits::eAllGraphics |
+			        vk::PipelineStageFlagBits::eComputeShader,
+			    shader_stages, vk::DependencyFlags {}, 0, nullptr, 1, &barrier, 0, nullptr);
+		}
+
+		for (uint32_t i = 0; i < program.info.images.size(); i++) {
+			auto& image   = m_context.GetTextureCache().GetImage(descriptors.images[i].image_id);
+			auto& binding = descriptors.images[i];
+			const auto&                 view = binding.desc.view_info;
+			const ImageSubresourceRange range {view.base_level, view.level_count, view.base_layer,
+			                                   view.layer_count};
+			const bool storage = binding.desc.type == TextureCache::BindingType::Storage;
+			if (image.info.data.Empty()) {
+				image.Transit(vk::ImageLayout::eGeneral,
+				              storage ? vk::AccessFlagBits2::eShaderRead |
+				                            vk::AccessFlagBits2::eShaderWrite
+				                      : vk::AccessFlagBits2::eShaderRead,
+				              range, vk_buffer);
+			} else if ((image.binding.force_general || image.binding.is_target) &&
+			           !image.info.IsDepth()) {
+				const vk::AccessFlags2 storage_access = image.binding.shader_write
+				                                            ? vk::AccessFlagBits2::eShaderWrite
+				                                            : vk::AccessFlags2 {};
+				image.Transit(vk::ImageLayout::eGeneral,
+				              vk::AccessFlagBits2::eShaderRead | storage_access |
+				                  vk::AccessFlagBits2::eColorAttachmentRead |
+				                  vk::AccessFlagBits2::eColorAttachmentWrite,
+				              {}, vk_buffer);
+			} else if (storage) {
+				image.Transit(vk::ImageLayout::eGeneral,
+				              vk::AccessFlagBits2::eShaderRead | vk::AccessFlagBits2::eShaderWrite,
+				              range, vk_buffer);
+			} else {
+				image.Transit(image.info.IsDepth() ? vk::ImageLayout::eDepthStencilReadOnlyOptimal
+				                                   : vk::ImageLayout::eShaderReadOnlyOptimal,
+				              vk::AccessFlagBits2::eShaderRead, range, vk_buffer);
+			}
+			binding.layout = image.backing.state.layout;
+		}
+
+		m_image_occurrences.assign(descriptors.images.size(), 0);
+		for (const auto& binding: program.bindings.descriptors) {
+			vk::WriteDescriptorSet write {};
+			write.sType          = vk::StructureType::eWriteDescriptorSet;
+			write.dstBinding     = ShaderRecompiler::IR::NativeBinding(program.stage, binding.kind);
+			write.descriptorType = NativeDescriptorType(binding.kind);
+			write.descriptorCount   = NativeDescriptorCount(binding);
+			const auto buffer_start = m_descriptor_buffers.size();
+			const auto image_start  = m_descriptor_images.size();
+			switch (binding.kind) {
+				case BindingKind::Buffers:
+					for (const auto resource: binding.resources) {
+						const auto& view = descriptors.buffers.at(resource);
+						EXIT_IF(view.buffer == nullptr);
+						m_descriptor_buffers.emplace_back(view.buffer, view.offset, view.range);
+					}
+					break;
+				case BindingKind::BdaPagetable:
+				case BindingKind::FaultBuffer: {
+					auto& cache = m_context.GetBufferCache();
+					const auto* bda_buffer = binding.kind == BindingKind::BdaPagetable
+					                             ? cache.GetBdaPageTableBuffer()
+					                             : cache.GetFaultBuffer();
+					m_descriptor_buffers.emplace_back(bda_buffer->Handle(), 0, bda_buffer->Size());
+					break;
+				}
+				case BindingKind::FlattenedSrt:
+				case BindingKind::UserData:
+				case BindingKind::Gds: {
+					const auto& view =
+					    binding.kind == BindingKind::FlattenedSrt ? descriptors.flattened_srt
+					    : binding.kind == BindingKind::UserData   ? descriptors.user_data
+					                                              : descriptors.gds;
+					EXIT_IF(view.buffer == nullptr);
+					m_descriptor_buffers.emplace_back(view.buffer, view.offset, view.range);
+					break;
+				}
+				case BindingKind::Samplers:
+					for (const auto resource: binding.resources) {
+						const auto sampler = descriptors.samplers.at(resource);
+						EXIT_IF(sampler == nullptr);
+						m_descriptor_images.emplace_back(sampler, nullptr,
+						                                 vk::ImageLayout::eUndefined);
+					}
+					break;
+				default:
+					for (const auto resource: binding.resources) {
+						m_descriptor_images.push_back(MakeImageInfo(
+						    descriptors.images.at(resource), m_image_occurrences.at(resource)++));
+					}
+					break;
+			}
+			if (m_descriptor_buffers.size() != buffer_start) {
+				write.pBufferInfo = m_descriptor_buffers.data() + buffer_start;
+			}
+			if (m_descriptor_images.size() != image_start) {
+				write.pImageInfo = m_descriptor_images.data() + image_start;
+			}
+			m_descriptor_writes.push_back(write);
+		}
+		for (uint32_t i = 0; i < descriptors.images.size(); i++) {
+			const auto expected =
+			    descriptors.images[i].mip_views.empty()
+			        ? 1u
+			        : static_cast<uint32_t>(descriptors.images[i].mip_views.size());
+			EXIT_IF(m_image_occurrences[i] != expected);
+		}
+
+		if (program.bindings.push_constant_size != 0) {
+			const auto offset = program.bindings.push_constant_offset;
+			EXIT_IF(offset % sizeof(uint32_t) != 0 ||
+			        program.bindings.push_constant_size !=
+			            prepared->user_data.size() * sizeof(uint32_t) ||
+			        offset + program.bindings.push_constant_size >
+			            ShaderRecompiler::IR::NativePushConstantSize);
+			std::copy(prepared->user_data.begin(), prepared->user_data.end(),
+			          m_push_constants.begin() + offset / sizeof(uint32_t));
+		}
 	}
-	if (program.bindings.push_constant_size != 0) {
-		EXIT_IF(program.bindings.push_constant_size !=
-		        prepared.user_data.size() * sizeof(uint32_t));
-		vk_buffer.pushConstants(layout, prepared.shader_stage,
-		                        program.bindings.push_constant_offset,
-		                        program.bindings.push_constant_size, prepared.user_data.data());
+	vk_buffer.pushConstants(pipeline.pipeline_layout, push_constant_stages, 0,
+	                        ShaderRecompiler::IR::NativePushConstantSize,
+	                        m_push_constants.data());
+
+	if (!m_descriptor_writes.empty()) {
+		EXIT_IF(pipeline.descriptor_set_layout == nullptr);
+		if (pipeline.uses_push_descriptors) {
+			vk_buffer.pushDescriptorSetKHR(pipeline_bind_point, pipeline.pipeline_layout, 0,
+			                               static_cast<uint32_t>(m_descriptor_writes.size()),
+			                               m_descriptor_writes.data());
+		} else {
+			const auto set = m_context.GetDescriptorHeap().Commit(pipeline.descriptor_set_layout);
+			for (auto& write: m_descriptor_writes) {
+				write.dstSet = set;
+			}
+			m_context.GetGraphics().device.updateDescriptorSets(
+			    static_cast<uint32_t>(m_descriptor_writes.size()), m_descriptor_writes.data(), 0,
+			    nullptr);
+			vk_buffer.bindDescriptorSets(pipeline_bind_point, pipeline.pipeline_layout, 0, 1, &set,
+			                             0, nullptr);
+		}
+	}
+	for (auto* prepared: prepared_bindings) {
+		prepared->committed = true;
 	}
-	prepared.committed = true;
 }
 
 } // namespace Libs::Graphics

src/graphics/host_gpu/renderer/pipeline/descriptors.h

@@ -2,6 +2,8 @@
 #define EMULATOR_SRC_GRAPHICS_HOST_GPU_RENDERER_DESCRIPTORS_H_
 
 #include "common/assert.h"
+#include "graphics/host_gpu/renderer/cache/bufferCache.h"
+#include "graphics/host_gpu/renderer/cache/textureCache.h"
 #include "graphics/host_gpu/renderer/image/image.h"
 #include "graphics/host_gpu/vulkanCommon.h"
 #include "graphics/shader/recompiler/ir/ShaderIR.h"
@@ -10,9 +12,54 @@
 #include <cstdint>
 #include <cstring>
 #include <type_traits>
+#include <vector>
 
 namespace Libs::Graphics {
 
+namespace ShaderRecompiler::IR {
+struct ResourceSnapshot;
+}
+
+struct BufferView {
+	vk::Buffer     buffer = nullptr;
+	vk::DeviceSize offset = 0;
+	vk::DeviceSize range  = VK_WHOLE_SIZE;
+};
+
+struct TextureBinding {
+	ImageId                    image_id;
+	vk::ImageView              image_view = nullptr;
+	TextureCache::ImageDesc    desc;
+	vk::ImageLayout            layout = vk::ImageLayout::eUndefined;
+	std::vector<vk::ImageView> mip_views;
+};
+
+struct NativeDescriptors {
+	std::vector<BufferView>     buffers;
+	std::vector<TextureBinding> images;
+	std::vector<vk::Sampler>    samplers;
+	BufferView                  gds;
+	BufferView                  flattened_srt;
+	BufferView                  user_data;
+};
+
+struct PreparedBindings {
+	const ShaderRecompiler::IR::Program*          program  = nullptr;
+	const ShaderRecompiler::IR::ResourceSnapshot* snapshot = nullptr;
+	NativeDescriptors                             resources;
+	std::vector<BufferId>                         buffer_ids;
+	std::vector<uint32_t>                         flattened_srt;
+	std::vector<uint32_t>                         user_data;
+	bool                                          committed = false;
+};
+
+[[nodiscard]] vk::DescriptorType
+NativeDescriptorType(ShaderRecompiler::IR::DescriptorBindingKind kind);
+[[nodiscard]] uint32_t
+NativeDescriptorCount(const ShaderRecompiler::IR::DescriptorBinding& binding);
+[[nodiscard]] vk::DescriptorImageInfo MakeImageInfo(const TextureBinding& texture,
+                                                    uint32_t              element = 0);
+
 template <typename T>
 [[nodiscard]] T DecodeNativeDescriptor(const ShaderRecompiler::IR::DescriptorValue& value) {
 	static_assert(std::is_trivially_copyable_v<T>);
@@ -34,11 +81,9 @@ ResolveTargetTextureView(const ShaderRecompiler::IR::ImageResource& resource,
                          Prospero::ImageType type, uint32_t base_layer, uint32_t image_layers);
 
 [[nodiscard]] bool IsSupportedDepthTargetDescriptor(const ShaderTextureResource& descriptor,
-                                                    const Image&                 image,
-                                                    bool r128 = false);
+                                                    const Image& image, bool r128 = false);
 [[nodiscard]] bool IsSupportedDepthTextureEncoding(const ShaderTextureResource& descriptor,
-                                                   const Image&                 image,
-                                                   bool r128 = false);
+                                                   const Image& image, bool r128 = false);
 [[nodiscard]] bool
 IsSupportedSampledVideoOutView(const ShaderRecompiler::IR::ImageResource& resource,
                                const ShaderTextureResource& descriptor, const Image& image);

src/graphics/host_gpu/renderer/pipeline/pipelineCache.cpp

@@ -1,6 +1,7 @@
 #include "graphics/host_gpu/renderer/pipeline/pipelineCache.h"
 
 #include "common/assert.h"
+#include "common/file.h"
 #include "common/logging/log.h"
 #include "common/profiler.h"
 #include "graphics/guest_gpu/hardwareContext.h"
@@ -10,16 +11,65 @@
 #include "graphics/host_gpu/renderer/image/imageView.h"
 #include "graphics/host_gpu/renderer/render.h"
 #include "graphics/host_gpu/renderer/renderContext.h"
+#include "graphics/shader/shaderCompiler.h"
+#include "kytyGitVersion.h"
+#include "loader/systemContent.h"
 
+#include <algorithm>
 #include <atomic>
+#include <cctype>
+#include <cstdio>
 #include <cstring>
+#include <limits>
 #include <span>
+#include <string_view>
 #include <utility>
+#include <vector>
+
+#include <fmt/format.h>
+#include <xxhash.h>
 
 namespace Libs::Graphics {
 
 namespace {
 
+std::string DriverCacheSignature(const vk::PhysicalDeviceProperties& properties) {
+	constexpr char hex[] = "0123456789abcdef";
+	std::string    uuid(VK_UUID_SIZE * 2, '0');
+	for (size_t i = 0; i < VK_UUID_SIZE; i++) {
+		uuid[i * 2]     = hex[properties.pipelineCacheUUID[i] >> 4u];
+		uuid[i * 2 + 1] = hex[properties.pipelineCacheUUID[i] & 0xfu];
+	}
+	return fmt::format("KytyPC1:{}:{:08x}:{:08x}:{:08x}:{}\n", KYTY_GIT_REVISION,
+	                   properties.vendorID, properties.deviceID, properties.driverVersion, uuid);
+}
+
+std::string PipelineCacheTitleId() {
+	std::string title_id;
+	if ((!Loader::SystemContentParamSfoGetString("TITLE_ID", &title_id) || title_id.empty()) &&
+	    (!Loader::SystemContentParamSfoGetString("CONTENT_ID", &title_id) || title_id.empty())) {
+		return {};
+	}
+	if (!std::ranges::all_of(title_id, [](unsigned char c) {
+		    return std::isalnum(c) != 0 || c == '-' || c == '_';
+	    })) {
+		return {};
+	}
+	return title_id;
+}
+
+template <typename... Args>
+void PipelineCacheLog(fmt::format_string<Args...> format, Args&&... args) {
+	auto message = fmt::format(format, std::forward<Args>(args)...);
+	message += '\n';
+	if (Log::GetDirection() != Log::Direction::Console) {
+		std::fwrite(message.data(), 1, message.size(), stdout);
+		std::fflush(stdout);
+	}
+	Log::Write(message);
+	Log::Flush();
+}
+
 void NormalizeStaticParamsForDynamicState(PipelineStaticParameters& static_params) {
 	static_params.viewport_scale[0]  = 0.5f;
 	static_params.viewport_scale[1]  = 0.5f;
@@ -36,16 +86,283 @@ void NormalizeStaticParamsForDynamicState(PipelineStaticParameters& static_param
 
 } // namespace
 
+struct PipelineCache::ProgramCache {
+	struct SourceKey {
+		ShaderType stage = ShaderType::Unknown;
+		uint64_t   hash  = 0;
+
+		bool operator==(const SourceKey&) const = default;
+	};
+
+	struct Permutation {
+		std::vector<uint32_t>                               static_key;
+		std::shared_ptr<const ShaderRecompiler::IR::Program> program;
+		ShaderProgram                                       handle;
+	};
+
+	struct SourceKeyHash {
+		std::size_t operator()(const SourceKey& key) const {
+			std::size_t hash = static_cast<std::size_t>(key.stage);
+			PipelineKeyHash::Mix(hash, static_cast<std::size_t>(key.hash));
+			if constexpr (sizeof(std::size_t) < sizeof(uint64_t)) {
+				PipelineKeyHash::Mix(hash, static_cast<std::size_t>(key.hash >> 32u));
+			}
+			return hash;
+		}
+	};
+
+	static uint64_t MixId(uint64_t hash, uint64_t value) {
+		return hash ^ (value + 0x9e3779b97f4a7c15ull + (hash << 6u) + (hash >> 2u));
+	}
+
+	static constexpr std::size_t MaxStaticKeyWords =
+	    5 + ShaderVertexInputInfo::RES_MAX * 17;
+
+	template <typename InputInfo>
+	ShaderProgram Get(const ShaderParams& params, InputInfo& input_info) {
+		constexpr ShaderType stage = [] {
+			if constexpr (std::is_same_v<InputInfo, ShaderVertexInputInfo>) {
+				return ShaderType::Vertex;
+			} else if constexpr (std::is_same_v<InputInfo, ShaderPixelInputInfo>) {
+				return ShaderType::Pixel;
+			} else {
+				static_assert(std::is_same_v<InputInfo, ShaderComputeInputInfo>);
+				return ShaderType::Compute;
+			}
+		}();
+
+		BuildStageStaticKey(input_info, key_scratch);
+		auto& permutations = programs[{stage, params.hash}];
+		for (const auto& permutation: permutations) {
+			if (permutation.static_key == key_scratch &&
+			    MaterializeProgram(permutation.program, params, input_info)) {
+				return permutation.handle;
+			}
+		}
+
+		const auto module = CompileProgram(device, params, input_info);
+		EXIT_IF(module == nullptr || !input_info.stage);
+		uint64_t id = MixId(MixId(params.hash, static_cast<uint64_t>(stage)), permutations.size());
+		if (id == 0) {
+			id = 1;
+		}
+		const ShaderProgram handle {.id = id, .module = module};
+		permutations.push_back({key_scratch, input_info.stage.program, handle});
+
+		std::printf("Num compiled %u shaders\n", ++num_compiled);
+		return handle;
+	}
+
+	explicit ProgramCache(vk::Device device): device(device) {
+		key_scratch.reserve(MaxStaticKeyWords);
+	}
+
+	std::unordered_map<SourceKey, std::vector<Permutation>, SourceKeyHash> programs;
+	std::vector<uint32_t> key_scratch;
+	vk::Device device;
+	uint32_t   num_compiled = 0;
+};
+
+PipelineCache::PipelineCache(GraphicContext& graphics)
+    : m_graphics(graphics), m_program_cache(std::make_unique<ProgramCache>(graphics.device)) {
+	EXIT_NOT_IMPLEMENTED(!Common::Thread::IsMainThread());
+	InitializeDriverCache();
+}
+
 PipelineCache::~PipelineCache() {
+	Save();
 	auto destroy = [this](const auto& pipelines) {
 		for (const auto& [key, pipeline]: pipelines) {
 			(void)key;
 			m_graphics.device.destroyPipeline(pipeline->pipeline, nullptr);
 			m_graphics.device.destroyPipelineLayout(pipeline->pipeline_layout, nullptr);
+			m_graphics.device.destroyDescriptorSetLayout(pipeline->descriptor_set_layout, nullptr);
 		}
 	};
 	destroy(m_graphics_pipelines);
 	destroy(m_compute_pipelines);
+	for (const auto& [key, permutations]: m_program_cache->programs) {
+		(void)key;
+		for (const auto& permutation: permutations) {
+			m_graphics.device.destroyShaderModule(permutation.handle.module, nullptr);
+		}
+	}
+	if (m_driver_cache != nullptr) {
+		m_graphics.device.destroyPipelineCache(m_driver_cache, nullptr);
+	}
+}
+
+void PipelineCache::InitializeDriverCache() {
+	const auto title_id = PipelineCacheTitleId();
+	if (title_id.empty()) {
+		return;
+	}
+	if (KYTY_BUILD != KYTY_BUILD_RELEASE) {
+		PipelineCacheLog("Vulkan pipeline cache: disabled (non-Release build)");
+		return;
+	}
+	const std::string_view git_hash     = KYTY_GIT_HASH;
+	const std::string_view git_revision = KYTY_GIT_REVISION;
+	if (git_hash == "unknown" || git_revision == "unknown") {
+		PipelineCacheLog("Vulkan pipeline cache: disabled (unknown git revision)");
+		return;
+	}
+	if (git_hash.ends_with("-dirty")) {
+		PipelineCacheLog("Vulkan pipeline cache: disabled (dirty build)");
+		return;
+	}
+
+	m_driver_cache_path = std::filesystem::path("_PipelineCache") / (title_id + ".bin");
+	const auto path         = Common::PathToString(m_driver_cache_path);
+	const bool cache_exists = Common::File::IsFileExisting(m_driver_cache_path);
+	if (cache_exists) {
+		PipelineCacheLog("Vulkan pipeline cache: loading {}", path);
+	} else {
+		PipelineCacheLog("Vulkan pipeline cache: initializing {}", path);
+	}
+	std::vector<uint8_t> initial_data;
+	if (cache_exists) {
+		Common::File file(m_driver_cache_path, Common::File::Mode::Read);
+		const auto   file_size = file.IsInvalid() ? 0 : file.Size();
+		const auto signature = DriverCacheSignature(m_graphics.GetPhysicalDeviceProperties());
+		if (file_size >= signature.size() + sizeof(uint64_t) &&
+		    file_size <= std::numeric_limits<uint32_t>::max()) {
+			std::string cached_signature(signature.size(), '\0');
+			uint64_t    payload_hash = 0;
+			initial_data.resize(file_size - signature.size() - sizeof(payload_hash));
+			uint32_t signature_read = 0;
+			uint32_t hash_read      = 0;
+			uint32_t payload_read   = 0;
+			file.Read(cached_signature.data(), static_cast<uint32_t>(cached_signature.size()),
+			          &signature_read);
+			file.Read(&payload_hash, sizeof(payload_hash), &hash_read);
+			file.Read(initial_data.data(), static_cast<uint32_t>(initial_data.size()), &payload_read);
+			file.Close();
+			if (signature_read != cached_signature.size() || hash_read != sizeof(payload_hash) ||
+			    payload_read != initial_data.size() || cached_signature != signature ||
+			    XXH3_64bits(initial_data.data(), initial_data.size()) != payload_hash) {
+				initial_data.clear();
+				PipelineCacheLog(
+				    "Vulkan pipeline cache: invalidating {} (driver, emulator, or data mismatch)",
+				    path);
+			}
+		} else {
+			file.Close();
+			PipelineCacheLog("Vulkan pipeline cache: invalidating {} (invalid file size)",
+			                 path);
+		}
+	}
+
+	vk::PipelineCacheCreateInfo create {};
+	create.sType           = vk::StructureType::ePipelineCacheCreateInfo;
+	create.initialDataSize = initial_data.size();
+	create.pInitialData    = initial_data.empty() ? nullptr : initial_data.data();
+	auto result = m_graphics.device.createPipelineCache(&create, nullptr, &m_driver_cache);
+	if (result != vk::Result::eSuccess && !initial_data.empty()) {
+		PipelineCacheLog("Vulkan pipeline cache: driver rejected {} ({}); starting empty", path,
+		                 VulkanToString(result));
+		initial_data.clear();
+		create.initialDataSize = 0;
+		create.pInitialData    = nullptr;
+		result = m_graphics.device.createPipelineCache(&create, nullptr, &m_driver_cache);
+	}
+	if (result != vk::Result::eSuccess) {
+		PipelineCacheLog("Vulkan pipeline cache: disabled ({})", VulkanToString(result));
+		m_driver_cache = nullptr;
+		return;
+	}
+	if (!initial_data.empty()) {
+		PipelineCacheLog("Vulkan pipeline cache: loaded {} bytes from {}", initial_data.size(),
+		                 path);
+	} else {
+		PipelineCacheLog("Vulkan pipeline cache: initialized empty");
+	}
+}
+
+void PipelineCache::Save() {
+	Common::LockGuard lock(m_mutex);
+	if (m_driver_cache == nullptr) {
+		return;
+	}
+
+	size_t               size = 0;
+	vk::Result           result;
+	std::vector<uint8_t> payload;
+	for (uint32_t attempt = 0; attempt < 3; attempt++) {
+		size   = 0;
+		result = m_graphics.device.getPipelineCacheData(m_driver_cache, &size, nullptr);
+		if (result != vk::Result::eSuccess || size == 0 ||
+		    size > std::numeric_limits<uint32_t>::max()) {
+			break;
+		}
+		payload.resize(size);
+		result = m_graphics.device.getPipelineCacheData(m_driver_cache, &size, payload.data());
+		if (result != vk::Result::eIncomplete) {
+			break;
+		}
+	}
+	if (result != vk::Result::eSuccess || size == 0 ||
+	    size > std::numeric_limits<uint32_t>::max()) {
+		PipelineCacheLog("Vulkan pipeline cache: save failed ({}, {} bytes)",
+		                 VulkanToString(result), size);
+		return;
+	}
+	payload.resize(size);
+	auto prefix = DriverCacheSignature(m_graphics.GetPhysicalDeviceProperties());
+	const auto payload_hash = XXH3_64bits(payload.data(), payload.size());
+	prefix.append(reinterpret_cast<const char*>(&payload_hash), sizeof(payload_hash));
+	if (!Common::File::CreateDirectories(m_driver_cache_path.parent_path())) {
+		PipelineCacheLog("Vulkan pipeline cache: failed to create cache directory");
+		return;
+	}
+	auto temp_path = m_driver_cache_path;
+	temp_path += ".tmp";
+	Common::File file;
+	uint32_t     prefix_written  = 0;
+	uint32_t     payload_written = 0;
+	if (file.Create(temp_path)) {
+		file.Write(prefix.data(), static_cast<uint32_t>(prefix.size()), &prefix_written);
+		file.Write(payload.data(), static_cast<uint32_t>(payload.size()), &payload_written);
+	}
+	const bool flushed = !file.IsInvalid() && file.Flush();
+	file.Close();
+	if (prefix_written != prefix.size() || payload_written != payload.size() || !flushed ||
+	    !Common::File::RenameFile(temp_path, m_driver_cache_path)) {
+		PipelineCacheLog("Vulkan pipeline cache: failed to write {}",
+		                 Common::PathToString(m_driver_cache_path));
+		return;
+	}
+	PipelineCacheLog("Vulkan pipeline cache: saved {} bytes to {}", payload.size(),
+	                 Common::PathToString(m_driver_cache_path));
+	m_graphics.device.destroyPipelineCache(m_driver_cache, nullptr);
+	m_driver_cache = nullptr;
+}
+
+ShaderProgram PipelineCache::GetVertexProgram(const HW::VertexShaderInfo& regs,
+                                              const HW::ShaderRegisters&  sh,
+                                              ShaderVertexInputInfo&      input_info) {
+	const auto params = PrepareProgram(regs, sh, input_info);
+	Common::LockGuard lock(m_mutex);
+	return m_program_cache->Get(params, input_info);
+}
+
+ShaderProgram PipelineCache::GetPixelProgram(
+    const HW::PixelShaderInfo& regs, const HW::ShaderRegisters& sh,
+    const ShaderVertexInputInfo&                        vertex_info,
+    std::span<const Prospero::ColorComponentMapping, 8> target_export_mapping,
+    ShaderPixelInputInfo&                               input_info) {
+	const auto params = PrepareProgram(regs, sh, vertex_info, target_export_mapping, input_info);
+	Common::LockGuard lock(m_mutex);
+	return m_program_cache->Get(params, input_info);
+}
+
+ShaderProgram PipelineCache::GetComputeProgram(const HW::ComputeShaderInfo& regs,
+                                               const HW::ShaderRegisters&   sh,
+                                               ShaderComputeInputInfo&      input_info) {
+	input_info.needs_lds_barriers = !m_graphics.compute_wave64_supported;
+	const auto params             = PrepareProgram(regs, sh, input_info);
+	Common::LockGuard lock(m_mutex);
+	return m_program_cache->Get(params, input_info);
 }
 
 bool PipelineStaticParameters::operator==(const PipelineStaticParameters& other) const noexcept {
@@ -53,24 +370,22 @@ bool PipelineStaticParameters::operator==(const PipelineStaticParameters& other)
 }
 
 PipelineCache::GraphicsPipeline& PipelineCache::CreateGraphicsPipeline(
-    RenderColorInfo* colors, uint32_t color_count, RenderDepthInfo& depth,
-    ShaderVertexInputInfo& vs_input_info, RenderCommandBuffer& command,
-    ShaderPixelInputInfo* ps_input_info, vk::PrimitiveTopology topology,
-    bool primitive_restart_enable, bool ps_active, std::span<const uint32_t> vs_spirv,
-    std::span<const uint32_t> ps_spirv) {
+    std::span<const RenderColorInfo> colors, const RenderDepthInfo& depth,
+    const ShaderVertexInputInfo& vs_input_info, CommandBuffer& command,
+    const ShaderPixelInputInfo* ps_input_info, vk::PrimitiveTopology topology,
+    bool primitive_restart_enable, const ShaderProgram& vertex_program,
+    const ShaderProgram& pixel_program) {
 	KYTY_PROFILER_BLOCK("PipelineCache::CreatePipeline(Gfx)", profiler::colors::DeepOrangeA200);
 
-	EXIT_IF(colors == nullptr);
-	EXIT_IF(color_count > RENDER_COLOR_ATTACHMENTS_MAX);
-	EXIT_IF(vs_spirv.empty());
-	EXIT_IF(ps_active && ps_spirv.empty());
+	EXIT_IF(colors.size() > RENDER_COLOR_ATTACHMENTS_MAX);
+	EXIT_IF(!vertex_program);
+	const bool ps_active = ps_input_info != nullptr;
+	EXIT_IF(ps_active && !pixel_program);
+	const auto color_count = static_cast<uint32_t>(colors.size());
 
 	Common::LockGuard lock(m_mutex);
-	auto&             ctx    = command.GetRegisters();
-	auto&             sh_ctx = command.GetShaders();
+	auto&             ctx = command.GetRegisters();
 
-	const auto&           vertex_info                              = sh_ctx.GetVs();
-	const auto&           ps_regs                                  = sh_ctx.GetPs();
 	const HW::BlendColor& bclr                                     = ctx.GetBlendColor();
 	uint32_t              color_mask[RENDER_COLOR_ATTACHMENTS_MAX] = {};
 	for (uint32_t i = 0; i < color_count; i++) {
@@ -81,11 +396,8 @@ PipelineCache::GraphicsPipeline& PipelineCache::CreateGraphicsPipeline(
 	}
 	const HW::ModeControl& mc = ctx.GetModeControl();
 
-	auto     vs_id = ShaderGetIdVS(vertex_info, vs_input_info, true);
-	ShaderId ps_id {};
-	if (ps_active) {
-		ps_id = ShaderGetIdPS(ps_regs, *ps_input_info, true);
-	}
+	const auto vs_id = vertex_program.id;
+	const auto ps_id = ps_active ? pixel_program.id : 0;
 
 	PipelineStaticParameters static_params {};
 	GraphicsPipeline         p {};
@@ -201,20 +513,17 @@ PipelineCache::GraphicsPipeline& PipelineCache::CreateGraphicsPipeline(
 		if (ps_active) {
 			ShaderDbgDumpInputInfo(*ps_input_info);
 		}
-		LOGF("PipelineTrace: shader binaries VS=0x%08" PRIx32 "/0x%08" PRIx32 " words=%" PRIu64
-		     " PS=0x%08" PRIx32 "/0x%08" PRIx32 " words=%" PRIu64 "\n",
-		     vs_id.hash0, vs_id.crc32, static_cast<uint64_t>(vs_spirv.size()), ps_id.hash0,
-		     ps_id.crc32, static_cast<uint64_t>(ps_spirv.size()));
+		LOGF("PipelineTrace: shader modules VS=%" PRIu64 " module=%p PS=%" PRIu64
+		     " module=%p\n",
+		     vs_id, static_cast<void*>(vertex_program.module), ps_id,
+		     static_cast<void*>(pixel_program.module));
 	}
 
 	auto cached = std::make_unique<GraphicsPipeline>(p);
-	LogPipelineTrace("CreatePipelineInternal begin", vs_id.hash0, vs_id.crc32, ps_id.hash0,
-	                 ps_id.crc32);
-	CreatePipelineInternal(m_graphics, m_descriptor_cache, *cached, rendering, vs_input_info,
-	                       vs_spirv, ps_input_info, ps_spirv, static_params, vs_id.hash0,
-	                       vs_id.crc32, ps_id.hash0, ps_id.crc32, ps_active);
-	LogPipelineTrace("CreatePipelineInternal done", vs_id.hash0, vs_id.crc32, ps_id.hash0,
-	                 ps_id.crc32);
+	LogPipelineTrace("CreatePipelineInternal begin", vs_id, ps_id);
+	CreatePipelineInternal(m_graphics, *cached, rendering, vs_input_info, vertex_program.module,
+	                       ps_input_info, pixel_program.module, static_params, m_driver_cache);
+	LogPipelineTrace("CreatePipelineInternal done", vs_id, ps_id);
 
 	EXIT_NOT_IMPLEMENTED(cached->pipeline == nullptr);
 	EXIT_NOT_IMPLEMENTED(cached->pipeline_layout == nullptr);
@@ -226,19 +535,16 @@ PipelineCache::GraphicsPipeline& PipelineCache::CreateGraphicsPipeline(
 }
 
 PipelineCache::ComputePipeline&
-PipelineCache::CreateComputePipeline(ShaderComputeInputInfo&      input_info,
-                                     const HW::ComputeShaderInfo& cs_regs,
-                                     std::span<const uint32_t>    cs_spirv) {
+PipelineCache::CreateComputePipeline(ShaderComputeInputInfo& input_info,
+                                     const ShaderProgram&    compute_program) {
 	KYTY_PROFILER_BLOCK("PipelineCache::CreatePipeline(Compute)", profiler::colors::RedA100);
 
-	EXIT_IF(cs_spirv.empty());
+	EXIT_IF(!compute_program);
 
 	Common::LockGuard lock(m_mutex);
 
-	auto cs_id = ShaderGetIdCS(cs_regs, input_info, true);
-
 	ComputePipeline p {};
-	p.cs_shader_id = cs_id;
+	p.cs_shader_id = compute_program.id;
 
 	ComputePipelineKey key {};
 	key.cs_shader_id = p.cs_shader_id;
@@ -252,7 +558,7 @@ PipelineCache::CreateComputePipeline(ShaderComputeInputInfo&      input_info,
 	}
 
 	auto cached = std::make_unique<ComputePipeline>(p);
-	CreatePipelineInternal(m_graphics, m_descriptor_cache, *cached, input_info, cs_spirv);
+	CreatePipelineInternal(m_graphics, *cached, input_info, compute_program.module, m_driver_cache);
 
 	EXIT_NOT_IMPLEMENTED(cached->pipeline == nullptr);
 	EXIT_NOT_IMPLEMENTED(cached->pipeline_layout == nullptr);

src/graphics/host_gpu/renderer/pipeline/pipelineCache.h

@@ -10,6 +10,7 @@
 #include "graphics/shader/shader.h"
 
 #include <cstddef>
+#include <filesystem>
 #include <memory>
 #include <span>
 #include <type_traits>
@@ -20,8 +21,7 @@ namespace Libs::Graphics {
 struct GraphicContext;
 struct RenderColorInfo;
 struct RenderDepthInfo;
-class RenderCommandBuffer;
-class DescriptorCache;
+class CommandBuffer;
 
 namespace HW {
 class Context;
@@ -96,44 +96,66 @@ struct PipelineRenderingState {
 	bool operator==(const PipelineRenderingState&) const = default;
 };
 
+struct ShaderProgram {
+	uint64_t         id     = 0;
+	vk::ShaderModule module = nullptr;
+
+	explicit operator bool() const {
+		return id != 0 && module != nullptr;
+	}
+};
+
 class PipelineCache {
 public:
-	PipelineCache(GraphicContext& graphics, DescriptorCache& descriptor_cache)
-	    : m_graphics(graphics), m_descriptor_cache(descriptor_cache) {
-		EXIT_NOT_IMPLEMENTED(!Common::Thread::IsMainThread());
-	}
+	explicit PipelineCache(GraphicContext& graphics);
 	~PipelineCache();
 	KYTY_CLASS_NO_COPY(PipelineCache);
+	void Save();
 
 	struct Pipeline {
-		vk::PipelineLayout pipeline_layout = nullptr;
-		vk::Pipeline       pipeline        = nullptr;
+		vk::PipelineLayout      pipeline_layout       = nullptr;
+		vk::Pipeline            pipeline              = nullptr;
+		vk::DescriptorSetLayout descriptor_set_layout = nullptr;
+		bool                    uses_push_descriptors = false;
 	};
 
 	struct GraphicsPipeline: Pipeline {
-		ShaderId vs_shader_id;
-		ShaderId ps_shader_id;
+		uint64_t vs_shader_id = 0;
+		uint64_t ps_shader_id = 0;
 	};
 
 	struct ComputePipeline: Pipeline {
-		ShaderId cs_shader_id;
+		uint64_t cs_shader_id = 0;
 	};
 
-	GraphicsPipeline&
-	CreateGraphicsPipeline(RenderColorInfo* colors, uint32_t color_count, RenderDepthInfo& depth,
-	                       ShaderVertexInputInfo& vs_input_info, RenderCommandBuffer& command,
-	                       ShaderPixelInputInfo* ps_input_info, vk::PrimitiveTopology topology,
-	                       bool primitive_restart_enable, bool ps_active,
-	                       std::span<const uint32_t> vs_spirv, std::span<const uint32_t> ps_spirv);
-	ComputePipeline& CreateComputePipeline(ShaderComputeInputInfo&      input_info,
-	                                       const HW::ComputeShaderInfo& cs_regs,
-	                                       std::span<const uint32_t>    cs_spirv);
+	ShaderProgram GetVertexProgram(const HW::VertexShaderInfo& regs,
+	                               const HW::ShaderRegisters&  sh,
+	                               ShaderVertexInputInfo&      input_info);
+	ShaderProgram
+	GetPixelProgram(const HW::PixelShaderInfo& regs, const HW::ShaderRegisters& sh,
+	                const ShaderVertexInputInfo&                        vertex_info,
+	                std::span<const Prospero::ColorComponentMapping, 8> target_export_mapping,
+	                ShaderPixelInputInfo&                               input_info);
+	ShaderProgram GetComputeProgram(const HW::ComputeShaderInfo& regs,
+	                                const HW::ShaderRegisters& sh,
+	                                ShaderComputeInputInfo& input_info);
+
+	GraphicsPipeline& CreateGraphicsPipeline(
+	    std::span<const RenderColorInfo> colors, const RenderDepthInfo& depth,
+	    const ShaderVertexInputInfo& vs_input_info, CommandBuffer& command,
+	    const ShaderPixelInputInfo* ps_input_info, vk::PrimitiveTopology topology,
+	    bool primitive_restart_enable, const ShaderProgram& vertex_program,
+	    const ShaderProgram& pixel_program);
+	ComputePipeline& CreateComputePipeline(ShaderComputeInputInfo& input_info,
+	                                       const ShaderProgram&    compute_program);
 
 private:
+	struct ProgramCache;
+
 	struct GraphicsPipelineKey {
 		PipelineRenderingState   rendering;
-		ShaderId                 vs_shader_id;
-		ShaderId                 ps_shader_id;
+		uint64_t                 vs_shader_id = 0;
+		uint64_t                 ps_shader_id = 0;
 		PipelineStaticParameters static_params;
 
 		bool operator==(const GraphicsPipelineKey& other) const {
@@ -143,7 +165,7 @@ private:
 	};
 
 	struct ComputePipelineKey {
-		ShaderId cs_shader_id;
+		uint64_t cs_shader_id = 0;
 
 		bool operator==(const ComputePipelineKey& other) const {
 			return cs_shader_id == other.cs_shader_id;
@@ -156,15 +178,6 @@ private:
 			        (hash >> 2u);
 		}
 
-		static void MixShaderId(std::size_t& hash, const ShaderId& id) {
-			Mix(hash, id.hash0);
-			Mix(hash, id.crc32);
-			Mix(hash, id.ids.size());
-			for (auto value: id.ids) {
-				Mix(hash, value);
-			}
-		}
-
 		static void MixStaticParams(std::size_t& hash, const PipelineStaticParameters& params) {
 			const auto* bytes = reinterpret_cast<const uint8_t*>(&params);
 			for (std::size_t i = 0; i < sizeof(params); i++) {
@@ -186,8 +199,8 @@ private:
 		std::size_t operator()(const GraphicsPipelineKey& key) const {
 			std::size_t hash = 0;
 			PipelineKeyHash::MixRendering(hash, key.rendering);
-			PipelineKeyHash::MixShaderId(hash, key.vs_shader_id);
-			PipelineKeyHash::MixShaderId(hash, key.ps_shader_id);
+			PipelineKeyHash::Mix(hash, key.vs_shader_id);
+			PipelineKeyHash::Mix(hash, key.ps_shader_id);
 			PipelineKeyHash::MixStaticParams(hash, key.static_params);
 			return hash;
 		}
@@ -196,34 +209,35 @@ private:
 	struct ComputePipelineKeyHash {
 		std::size_t operator()(const ComputePipelineKey& key) const {
 			std::size_t hash = 0;
-			PipelineKeyHash::MixShaderId(hash, key.cs_shader_id);
+			PipelineKeyHash::Mix(hash, key.cs_shader_id);
 			return hash;
 		}
 	};
 
-	GraphicContext&  m_graphics;
-	DescriptorCache& m_descriptor_cache;
+	GraphicContext&                m_graphics;
+	std::unique_ptr<ProgramCache> m_program_cache;
+	vk::PipelineCache             m_driver_cache = nullptr;
+	std::filesystem::path         m_driver_cache_path;
 	std::unordered_map<GraphicsPipelineKey, std::unique_ptr<GraphicsPipeline>,
 	                   GraphicsPipelineKeyHash>
 	    m_graphics_pipelines;
 	std::unordered_map<ComputePipelineKey, std::unique_ptr<ComputePipeline>, ComputePipelineKeyHash>
 	              m_compute_pipelines;
 	Common::Mutex m_mutex;
+
+	void InitializeDriverCache();
 };
 
-void LogPipelineTrace(const char* phase, uint32_t vs_hash0, uint32_t vs_crc32, uint32_t ps_hash0,
-                      uint32_t ps_crc32);
+void LogPipelineTrace(const char* phase, uint64_t vertex_program_id, uint64_t pixel_program_id);
 void CreatePipelineInternal(
-    GraphicContext& graphics, DescriptorCache& descriptor_cache,
-    PipelineCache::GraphicsPipeline& pipeline, const PipelineRenderingState& rendering,
-    const ShaderVertexInputInfo& vs_input_info, std::span<const uint32_t> vs_shader,
-    const ShaderPixelInputInfo* ps_input_info, std::span<const uint32_t> ps_shader,
-    const PipelineStaticParameters& static_params, uint32_t vs_hash0, uint32_t vs_crc32,
-    uint32_t ps_hash0, uint32_t ps_crc32, bool ps_active);
-void CreatePipelineInternal(GraphicContext& graphics, DescriptorCache& descriptor_cache,
-                            PipelineCache::ComputePipeline& pipeline,
-                            const ShaderComputeInputInfo&   input_info,
-                            std::span<const uint32_t>       cs_shader);
+    GraphicContext& graphics, PipelineCache::GraphicsPipeline& pipeline,
+    const PipelineRenderingState& rendering, const ShaderVertexInputInfo& vs_input_info,
+    vk::ShaderModule vertex_module, const ShaderPixelInputInfo* ps_input_info,
+	vk::ShaderModule pixel_module, const PipelineStaticParameters& static_params,
+	vk::PipelineCache driver_cache);
+void CreatePipelineInternal(GraphicContext& graphics, PipelineCache::ComputePipeline& pipeline,
+                            const ShaderComputeInputInfo& input_info,
+	vk::ShaderModule compute_module, vk::PipelineCache driver_cache);
 
 } // namespace Libs::Graphics
 

src/graphics/host_gpu/renderer/pipeline/shaderSubgroup.cpp

@@ -1,58 +0,0 @@
-#include "graphics/host_gpu/renderer/pipeline/shaderSubgroup.h"
-
-#include "graphics/shader/recompiler/backend/spirv/SpirvEmitter.h"
-
-namespace Libs::Graphics {
-
-ShaderSubgroupCapabilities::ShaderSubgroupCapabilities(const GraphicContext& graphics)
-    : subgroup_size(graphics.subgroup_size), min_subgroup_size(graphics.min_subgroup_size),
-      max_subgroup_size(graphics.max_subgroup_size),
-      required_subgroup_size_stages(graphics.required_subgroup_size_stages),
-      subgroup_size_control_enabled(graphics.subgroup_size_control_enabled) {}
-
-ShaderLaneMaskMode SelectGraphicsLaneMaskMode(uint32_t guest_wave_size) {
-	// A 64-wide guest wave can't assume the host GPU's lane layout matches
-	// GCN's. AMD RDNA also reports a 64-wide subgroup but uses a different
-	// lane/pixel layout, which corrupted rendering under "native wave" mode.
-	// Per-invocation emulation works regardless of the host's lane layout.
-	if (guest_wave_size == 64u) {
-		return ShaderLaneMaskMode::PerInvocation;
-	}
-	return ShaderLaneMaskMode::NativeWave;
-}
-
-ShaderSubgroupConfiguration ConfigureShaderSubgroup(const ShaderSubgroupCapabilities& capabilities,
-                                                    vk::ShaderStageFlagBits           stage,
-                                                    const ShaderRecompiler::IR::Program& program) {
-	const auto guest_wave_size = program.wave_size;
-	if ((guest_wave_size != 32u && guest_wave_size != 64u) ||
-	    !program.spirv_requirements.has_value()) {
-		return {};
-	}
-	if (stage != vk::ShaderStageFlagBits::eCompute) {
-		const auto expected = SelectGraphicsLaneMaskMode(guest_wave_size);
-		if (program.lane_mask_mode != expected || (capabilities.subgroup_size != guest_wave_size &&
-		                                           expected != ShaderLaneMaskMode::PerInvocation)) {
-			return {};
-		}
-		return {expected == ShaderLaneMaskMode::NativeWave
-		            ? ShaderSubgroupMode::Natural
-		            : ShaderSubgroupMode::PerInvocationGraphics,
-		        0};
-	}
-	if (program.lane_mask_mode != ShaderLaneMaskMode::NativeWave) {
-		return {};
-	}
-	if (capabilities.subgroup_size == guest_wave_size) {
-		return {ShaderSubgroupMode::Natural, 0};
-	}
-	if (capabilities.subgroup_size_control_enabled &&
-	    (capabilities.required_subgroup_size_stages & stage) &&
-	    guest_wave_size >= capabilities.min_subgroup_size &&
-	    guest_wave_size <= capabilities.max_subgroup_size) {
-		return {ShaderSubgroupMode::Controlled, guest_wave_size};
-	}
-	return {ShaderSubgroupMode::FlattenedMasks, 0};
-}
-
-} // namespace Libs::Graphics

src/graphics/host_gpu/renderer/pipeline/shaderSubgroup.h

@@ -1,40 +0,0 @@
-#ifndef EMULATOR_SRC_GRAPHICS_HOST_GPU_RENDERER_SHADERSUBGROUP_H_
-#define EMULATOR_SRC_GRAPHICS_HOST_GPU_RENDERER_SHADERSUBGROUP_H_
-
-#include "graphics/host_gpu/graphicContext.h"
-#include "graphics/shader/recompiler/ir/ShaderIR.h"
-
-namespace Libs::Graphics {
-
-enum class ShaderSubgroupMode {
-	Natural,
-	Controlled,
-	PerInvocationGraphics,
-	FlattenedMasks,
-	Unsupported
-};
-
-struct ShaderSubgroupConfiguration {
-	ShaderSubgroupMode mode          = ShaderSubgroupMode::Unsupported;
-	uint32_t           required_size = 0;
-};
-
-struct ShaderSubgroupCapabilities {
-	explicit ShaderSubgroupCapabilities(const GraphicContext& graphics);
-
-	uint32_t             subgroup_size     = 0;
-	uint32_t             min_subgroup_size = 0;
-	uint32_t             max_subgroup_size = 0;
-	vk::ShaderStageFlags required_subgroup_size_stages;
-	bool                 subgroup_size_control_enabled = false;
-};
-
-ShaderLaneMaskMode SelectGraphicsLaneMaskMode(uint32_t guest_wave_size);
-
-ShaderSubgroupConfiguration ConfigureShaderSubgroup(const ShaderSubgroupCapabilities& capabilities,
-                                                    vk::ShaderStageFlagBits           stage,
-                                                    const ShaderRecompiler::IR::Program& program);
-
-} // namespace Libs::Graphics
-
-#endif // EMULATOR_SRC_GRAPHICS_HOST_GPU_RENDERER_SHADERSUBGROUP_H_

src/graphics/host_gpu/renderer/pipeline/shaders.cpp

@@ -6,9 +6,8 @@
 #include "graphics/guest_gpu/gpu_defs.h"
 #include "graphics/host_gpu/graphicContext.h"
 #include "graphics/host_gpu/renderer/debug.h"
-#include "graphics/host_gpu/renderer/pipeline/descriptorCache.h"
+#include "graphics/host_gpu/renderer/pipeline/descriptors.h"
 #include "graphics/host_gpu/renderer/pipeline/pipelineCache.h"
-#include "graphics/host_gpu/renderer/pipeline/shaderSubgroup.h"
 #include "graphics/host_gpu/renderer/render.h"
 #include "graphics/host_gpu/renderer/renderContext.h"
 #include "graphics/host_gpu/renderer/renderTarget.h"
@@ -20,6 +19,7 @@
 #include <algorithm>
 #include <limits>
 #include <span>
+#include <vector>
 
 namespace Libs::Graphics {
 
@@ -389,116 +389,53 @@ static vk::BlendOp GetBlendOp(uint32_t op) {
 	return vk::BlendOp::eAdd;
 }
 
-static void CreateLayout(DescriptorCache&                   descriptor_cache,
-                         std::span<vk::DescriptorSetLayout> set_layouts, uint32_t& set_layouts_num,
-                         std::span<vk::PushConstantRange>     push_constant_info,
-                         uint32_t&                            push_constant_info_num,
-                         const ShaderRecompiler::IR::Program& program,
-                         vk::ShaderStageFlags vk_stage, DescriptorCache::Stage stage) {
-	const auto& bindings        = program.bindings;
-	const bool  need_descriptor = !bindings.descriptors.empty();
-
-	if (bindings.push_constant_size != 0) {
-		auto index = push_constant_info_num;
-
-		push_constant_info[index].stageFlags = vk_stage;
-		push_constant_info[index].offset     = bindings.push_constant_offset;
-		push_constant_info[index].size       = bindings.push_constant_size;
-		push_constant_info_num++;
-	}
-
-	if (need_descriptor) {
-		EXIT_IF(bindings.descriptor_set != set_layouts_num);
-
-		set_layouts[set_layouts_num] = descriptor_cache.GetDescriptorSetLayout(stage, program);
-		set_layouts_num++;
+static void AddLayoutBindings(std::vector<vk::DescriptorSetLayoutBinding>& descriptor_bindings,
+                              const ShaderRecompiler::IR::Program& program,
+                              vk::ShaderStageFlagBits              stage) {
+	for (const auto& binding: program.bindings.descriptors) {
+		descriptor_bindings.push_back(
+		    {ShaderRecompiler::IR::NativeBinding(program.stage, binding.kind),
+		     NativeDescriptorType(binding.kind), NativeDescriptorCount(binding), stage, nullptr});
 	}
 }
 
-static void ConfigureSubgroupSize(const GraphicContext& graphics, vk::ShaderStageFlagBits vk_stage,
-                                  const ShaderRecompiler::IR::Program&                   program,
-                                  vk::PipelineShaderStageRequiredSubgroupSizeCreateInfo& required,
-                                  vk::PipelineShaderStageCreateInfo&                     stage) {
-	const auto config =
-	    ConfigureShaderSubgroup(ShaderSubgroupCapabilities {graphics}, vk_stage, program);
-	switch (config.mode) {
-		case ShaderSubgroupMode::Natural: return;
-		case ShaderSubgroupMode::PerInvocationGraphics: {
-			static std::atomic<uint32_t> log_count {0};
-			if (log_count.fetch_add(1, std::memory_order_relaxed) < 16) {
-				LOGF("Vulkan running %u-lane graphics shader stage 0x%08x with per-invocation "
-				     "EXEC/VCC on the host-native subgroup\n",
-				     program.wave_size, static_cast<uint32_t>(vk_stage));
-			}
-			return;
-		}
-		case ShaderSubgroupMode::FlattenedMasks: {
-			static std::atomic<uint32_t> log_count {0};
-			if (log_count.fetch_add(1, std::memory_order_relaxed) < 16) {
-				LOGF("Vulkan running %u-lane shader stage 0x%08x with flattened per-invocation "
-				     "EXEC/VCC\n",
-				     program.wave_size, static_cast<uint32_t>(vk_stage));
-			}
-			return;
-		}
-		case ShaderSubgroupMode::Controlled: break;
-		case ShaderSubgroupMode::Unsupported:
-		default:
-			EXIT("Vulkan cannot run %u-lane shader stage 0x%08x: default=%u min=%u max=%u "
-			     "controlled_stages=0x%08x\n",
-			     program.wave_size, static_cast<uint32_t>(vk_stage), graphics.subgroup_size,
-			     graphics.min_subgroup_size, graphics.max_subgroup_size,
-			     static_cast<vk::ShaderStageFlags::MaskType>(
-			         graphics.required_subgroup_size_stages));
+static void CreateDescriptorLayout(GraphicContext& graphics, PipelineCache::Pipeline& pipeline,
+                                   std::span<const vk::DescriptorSetLayoutBinding> bindings) {
+	uint32_t descriptor_count = 0;
+	for (const auto& binding: bindings) {
+		descriptor_count += binding.descriptorCount;
 	}
+	pipeline.uses_push_descriptors = descriptor_count <= graphics.max_push_descriptors;
 
-	required.sType = vk::StructureType::ePipelineShaderStageRequiredSubgroupSizeCreateInfo;
-	required.requiredSubgroupSize = config.required_size;
-	stage.pNext                   = &required;
+	vk::DescriptorSetLayoutCreateInfo create {};
+	create.sType        = vk::StructureType::eDescriptorSetLayoutCreateInfo;
+	create.flags        = pipeline.uses_push_descriptors
+	                          ? vk::DescriptorSetLayoutCreateFlagBits::ePushDescriptorKHR
+	                          : vk::DescriptorSetLayoutCreateFlags {};
+	create.bindingCount = static_cast<uint32_t>(bindings.size());
+	create.pBindings    = bindings.data();
+	EXIT_IF(graphics.device.createDescriptorSetLayout(
+	            &create, nullptr, &pipeline.descriptor_set_layout) != vk::Result::eSuccess);
 }
 
 // NOLINTNEXTLINE(readability-function-cognitive-complexity)
 void CreatePipelineInternal(
-    GraphicContext& graphics, DescriptorCache& descriptor_cache,
-    PipelineCache::GraphicsPipeline& pipeline, const PipelineRenderingState& rendering,
-    const ShaderVertexInputInfo& vs_input_info, std::span<const uint32_t> vs_shader,
-    const ShaderPixelInputInfo* ps_input_info, std::span<const uint32_t> ps_shader,
-    const PipelineStaticParameters& static_params, uint32_t vs_hash0, uint32_t vs_crc32,
-    uint32_t ps_hash0, uint32_t ps_crc32, bool ps_active) {
-	EXIT_IF(ps_active && ps_input_info == nullptr);
+    GraphicContext& graphics, PipelineCache::GraphicsPipeline& pipeline,
+    const PipelineRenderingState& rendering, const ShaderVertexInputInfo& vs_input_info,
+    vk::ShaderModule vertex_module, const ShaderPixelInputInfo* ps_input_info,
+	vk::ShaderModule pixel_module, const PipelineStaticParameters& static_params,
+	vk::PipelineCache driver_cache) {
+	const bool ps_active = ps_input_info != nullptr;
+	EXIT_IF(vertex_module == nullptr || (ps_active && pixel_module == nullptr));
 
 	const bool rect_list = static_params.topology == vk::PrimitiveTopology::ePatchList;
 
-	vk::ShaderModule vert_shader_module         = nullptr;
 	vk::ShaderModule tess_control_shader_module = nullptr;
 	vk::ShaderModule tess_eval_shader_module    = nullptr;
-	vk::ShaderModule frag_shader_module         = nullptr;
 
 	vk::ShaderModuleCreateInfo create_info {};
-
 	create_info.sType = vk::StructureType::eShaderModuleCreateInfo;
-	create_info.pNext = nullptr;
-	create_info.flags = {};
-
-	create_info.codeSize = vs_shader.size() * 4;
-	create_info.pCode    = vs_shader.data();
-	auto result = graphics.device.createShaderModule(&create_info, nullptr, &vert_shader_module);
-	if (graphics_debug_dump_enabled()) {
-		LOGF("PipelineTrace: vkCreateShaderModule VS done result=%s module=%p\n",
-		     VulkanToString(result).c_str(), static_cast<void*>(vert_shader_module));
-	}
-	EXIT_NOT_IMPLEMENTED(result != vk::Result::eSuccess);
-
-	if (ps_active) {
-		create_info.codeSize = ps_shader.size() * 4;
-		create_info.pCode    = ps_shader.data();
-		result = graphics.device.createShaderModule(&create_info, nullptr, &frag_shader_module);
-		if (graphics_debug_dump_enabled()) {
-			LOGF("PipelineTrace: vkCreateShaderModule PS done result=%s module=%p\n",
-			     VulkanToString(result).c_str(), static_cast<void*>(frag_shader_module));
-		}
-		EXIT_NOT_IMPLEMENTED(result != vk::Result::eSuccess);
-	}
+	vk::Result result {};
 	if (rect_list) {
 		const auto shaders =
 		    BuildRectListShaders(vs_input_info, ps_active ? ps_input_info : nullptr);
@@ -523,40 +460,27 @@ void CreatePipelineInternal(
 		EXIT_NOT_IMPLEMENTED(result != vk::Result::eSuccess);
 	}
 
-	EXIT_NOT_IMPLEMENTED(vert_shader_module == nullptr);
 	EXIT_NOT_IMPLEMENTED(
 	    rect_list && (tess_control_shader_module == nullptr || tess_eval_shader_module == nullptr));
-	EXIT_NOT_IMPLEMENTED(ps_active && frag_shader_module == nullptr);
 
-	vk::PipelineShaderStageCreateInfo                     vert_shader_stage_info {};
-	vk::PipelineShaderStageRequiredSubgroupSizeCreateInfo vert_subgroup_size {};
+	vk::PipelineShaderStageCreateInfo vert_shader_stage_info {};
 
 	vert_shader_stage_info.sType               = vk::StructureType::ePipelineShaderStageCreateInfo;
 	vert_shader_stage_info.pNext               = nullptr;
 	vert_shader_stage_info.flags               = {};
 	vert_shader_stage_info.stage               = vk::ShaderStageFlagBits::eVertex;
-	vert_shader_stage_info.module              = vert_shader_module;
+	vert_shader_stage_info.module              = vertex_module;
 	vert_shader_stage_info.pName               = "main";
 	vert_shader_stage_info.pSpecializationInfo = nullptr;
-	EXIT_IF(!vs_input_info.stage);
-	ConfigureSubgroupSize(graphics, vk::ShaderStageFlagBits::eVertex, *vs_input_info.stage.program,
-	                      vert_subgroup_size, vert_shader_stage_info);
 
-	vk::PipelineShaderStageCreateInfo                     frag_shader_stage_info {};
-	vk::PipelineShaderStageRequiredSubgroupSizeCreateInfo frag_subgroup_size {};
+	vk::PipelineShaderStageCreateInfo frag_shader_stage_info {};
 	frag_shader_stage_info.sType               = vk::StructureType::ePipelineShaderStageCreateInfo;
 	frag_shader_stage_info.pNext               = nullptr;
 	frag_shader_stage_info.flags               = {};
 	frag_shader_stage_info.stage               = vk::ShaderStageFlagBits::eFragment;
-	frag_shader_stage_info.module              = frag_shader_module;
+	frag_shader_stage_info.module              = pixel_module;
 	frag_shader_stage_info.pName               = "main";
 	frag_shader_stage_info.pSpecializationInfo = nullptr;
-	if (ps_active) {
-		EXIT_IF(!ps_input_info->stage);
-		ConfigureSubgroupSize(graphics, vk::ShaderStageFlagBits::eFragment,
-		                      *ps_input_info->stage.program, frag_subgroup_size,
-		                      frag_shader_stage_info);
-	}
 
 	vk::PipelineShaderStageCreateInfo tess_control_shader_stage_info {};
 	tess_control_shader_stage_info.sType  = vk::StructureType::ePipelineShaderStageCreateInfo;
@@ -599,9 +523,10 @@ void CreatePipelineInternal(
 
 			uint32_t attr_size       = 4;
 			auto     registers_num   = vs_input_info.resources_dst[index].registers_num;
-			auto     used_components = (vs_input_info.resource_fetch_components[index] > 0
-			                                ? vs_input_info.resource_fetch_components[index]
-			                                : registers_num);
+			const auto compiled_components =
+			    vs_input_info.stage.program->info.vertex_fetch_components[index];
+			auto used_components =
+			    compiled_components > 0 ? static_cast<int>(compiled_components) : registers_num;
 			GetInputFormat(vs_input_info.resources[index], input_attr[index].format, attr_size,
 			               static_cast<uint32_t>(used_components));
 
@@ -877,40 +802,36 @@ void CreatePipelineInternal(
 	color_blending.blendConstants[2] = static_params.blend_color_blue;
 	color_blending.blendConstants[3] = static_params.blend_color_alpha;
 
-	vk::DescriptorSetLayout set_layouts[2]  = {};
-	uint32_t                set_layouts_num = 0;
-
-	vk::PushConstantRange push_constant_info[2];
-	uint32_t              push_constant_info_num = 0;
-
 	EXIT_IF(!vs_input_info.stage);
-	CreateLayout(descriptor_cache, set_layouts, set_layouts_num, push_constant_info,
-	             push_constant_info_num, *vs_input_info.stage.program,
-	             vk::ShaderStageFlagBits::eVertex, DescriptorCache::Stage::Vertex);
+	std::vector<vk::DescriptorSetLayoutBinding> descriptor_bindings;
+	AddLayoutBindings(descriptor_bindings, *vs_input_info.stage.program,
+	                  vk::ShaderStageFlagBits::eVertex);
 	if (ps_active) {
 		EXIT_IF(!ps_input_info->stage);
-		CreateLayout(descriptor_cache, set_layouts, set_layouts_num, push_constant_info,
-		             push_constant_info_num, *ps_input_info->stage.program,
-		             vk::ShaderStageFlagBits::eFragment, DescriptorCache::Stage::Pixel);
+		AddLayoutBindings(descriptor_bindings, *ps_input_info->stage.program,
+		                  vk::ShaderStageFlagBits::eFragment);
 	}
+	CreateDescriptorLayout(graphics, pipeline, descriptor_bindings);
+	constexpr auto GraphicsStages =
+	    vk::ShaderStageFlagBits::eVertex | vk::ShaderStageFlagBits::eFragment;
+	const vk::PushConstantRange push_constants {
+	    GraphicsStages, 0, ShaderRecompiler::IR::NativePushConstantSize};
 
 	vk::PipelineLayoutCreateInfo pipeline_layout_info {};
 	pipeline_layout_info.sType                  = vk::StructureType::ePipelineLayoutCreateInfo;
 	pipeline_layout_info.pNext                  = nullptr;
 	pipeline_layout_info.flags                  = {};
-	pipeline_layout_info.setLayoutCount         = set_layouts_num;
-	pipeline_layout_info.pSetLayouts            = (set_layouts_num > 0 ? set_layouts : nullptr);
-	pipeline_layout_info.pushConstantRangeCount = push_constant_info_num;
-	pipeline_layout_info.pPushConstantRanges =
-	    push_constant_info_num > 0 ? push_constant_info : nullptr;
+	pipeline_layout_info.setLayoutCount         = 1;
+	pipeline_layout_info.pSetLayouts            = &pipeline.descriptor_set_layout;
+	pipeline_layout_info.pushConstantRangeCount = 1;
+	pipeline_layout_info.pPushConstantRanges    = &push_constants;
 
 	EXIT_IF(pipeline.pipeline_layout != nullptr);
 
 	if (graphics_debug_dump_enabled()) {
-		LOGF("PipelineTrace: vkCreatePipelineLayout begin VS=0x%08" PRIx32 "/0x%08" PRIx32
-		     " PS=0x%08" PRIx32 "/0x%08" PRIx32 " set_layouts=%" PRIu32 " push_constants=%" PRIu32
-		     "\n",
-		     vs_hash0, vs_crc32, ps_hash0, ps_crc32, set_layouts_num, push_constant_info_num);
+		LOGF("PipelineTrace: vkCreatePipelineLayout begin VS=%" PRIu64 " PS=%" PRIu64
+		     " set_layouts=1 push_constants=%" PRIu32 "\n",
+		     pipeline.vs_shader_id, pipeline.ps_shader_id, 1u);
 	}
 	result = graphics.device.createPipelineLayout(&pipeline_layout_info, nullptr,
 	                                              &pipeline.pipeline_layout);
@@ -1009,17 +930,18 @@ void CreatePipelineInternal(
 	EXIT_IF(pipeline.pipeline != nullptr);
 
 	if (graphics_debug_dump_enabled()) {
-		LOGF("PipelineTrace: vkCreateGraphicsPipelines begin VS=0x%08" PRIx32 "/0x%08" PRIx32
-		     " PS=0x%08" PRIx32 "/0x%08" PRIx32 " topology=%" PRIu32 " color_mask=0x%08" PRIx32
+		LOGF("PipelineTrace: vkCreateGraphicsPipelines begin VS=%" PRIu64 " PS=%" PRIu64
+		     " topology=%" PRIu32 " color_mask=0x%08" PRIx32
 		     " depth=%s blend=%s dyn_states=%" PRIu32 " viewport=%f,%f %fx%f"
 		     " scissor=%d,%d %ux%u\n",
-		     vs_hash0, vs_crc32, ps_hash0, ps_crc32, static_cast<uint32_t>(static_params.topology),
+		     pipeline.vs_shader_id, pipeline.ps_shader_id,
+		     static_cast<uint32_t>(static_params.topology),
 		     static_params.color_mask[0], (static_params.with_depth ? "true" : "false"),
 		     (static_params.blend_enable[0] ? "true" : "false"), dynamic_states_count, viewport.x,
 		     viewport.y, viewport.width, viewport.height, scissor.offset.x, scissor.offset.y,
 		     scissor.extent.width, scissor.extent.height);
 	}
-	result = graphics.device.createGraphicsPipelines(nullptr, 1, &pipeline_info, nullptr,
+	result = graphics.device.createGraphicsPipelines(driver_cache, 1, &pipeline_info, nullptr,
 	                                                 &pipeline.pipeline);
 	if (graphics_debug_dump_enabled()) {
 		LOGF("PipelineTrace: vkCreateGraphicsPipelines done result=%s pipeline=%p\n",
@@ -1029,40 +951,19 @@ void CreatePipelineInternal(
 
 	EXIT_NOT_IMPLEMENTED(pipeline.pipeline == nullptr);
 
-	if (frag_shader_module != nullptr) {
-		graphics.device.destroyShaderModule(frag_shader_module, nullptr);
-	}
 	if (tess_control_shader_module != nullptr) {
 		graphics.device.destroyShaderModule(tess_control_shader_module, nullptr);
 	}
 	if (tess_eval_shader_module != nullptr) {
 		graphics.device.destroyShaderModule(tess_eval_shader_module, nullptr);
 	}
-	graphics.device.destroyShaderModule(vert_shader_module, nullptr);
 }
 
 // NOLINTNEXTLINE(readability-function-cognitive-complexity)
-void CreatePipelineInternal(GraphicContext& graphics, DescriptorCache& descriptor_cache,
-                            PipelineCache::ComputePipeline& pipeline,
-                            const ShaderComputeInputInfo&   input_info,
-                            std::span<const uint32_t>       cs_shader) {
-	vk::ShaderModule comp_shader_module = nullptr;
-
-	vk::ShaderModuleCreateInfo create_info {};
-
-	create_info.sType    = vk::StructureType::eShaderModuleCreateInfo;
-	create_info.pNext    = nullptr;
-	create_info.flags    = {};
-	create_info.codeSize = cs_shader.size() * 4;
-	create_info.pCode    = cs_shader.data();
-	LOGF("PipelineTrace: vkCreateShaderModule CS begin words=%" PRIu64 "\n",
-	     static_cast<uint64_t>(cs_shader.size()));
-	auto result = graphics.device.createShaderModule(&create_info, nullptr, &comp_shader_module);
-	LOGF("PipelineTrace: vkCreateShaderModule CS done result=%s module=%p\n",
-	     VulkanToString(result).c_str(), static_cast<void*>(comp_shader_module));
-	EXIT_NOT_IMPLEMENTED(result != vk::Result::eSuccess);
-
-	EXIT_NOT_IMPLEMENTED(comp_shader_module == nullptr);
+void CreatePipelineInternal(GraphicContext& graphics, PipelineCache::ComputePipeline& pipeline,
+                            const ShaderComputeInputInfo& input_info,
+	vk::ShaderModule compute_module, vk::PipelineCache driver_cache) {
+	EXIT_IF(compute_module == nullptr);
 
 	vk::PipelineShaderStageCreateInfo                     comp_shader_stage_info {};
 	vk::PipelineShaderStageRequiredSubgroupSizeCreateInfo comp_subgroup_size {};
@@ -1070,40 +971,41 @@ void CreatePipelineInternal(GraphicContext& graphics, DescriptorCache& descripto
 	comp_shader_stage_info.pNext               = nullptr;
 	comp_shader_stage_info.flags               = {};
 	comp_shader_stage_info.stage               = vk::ShaderStageFlagBits::eCompute;
-	comp_shader_stage_info.module              = comp_shader_module;
+	comp_shader_stage_info.module              = compute_module;
 	comp_shader_stage_info.pName               = "main";
 	comp_shader_stage_info.pSpecializationInfo = nullptr;
 	EXIT_IF(!input_info.stage);
-	ConfigureSubgroupSize(graphics, vk::ShaderStageFlagBits::eCompute, *input_info.stage.program,
-	                      comp_subgroup_size, comp_shader_stage_info);
+	const auto wave_size = input_info.stage.program->wave_size;
+	if (graphics.compute_subgroup_size_control_enabled &&
+	    wave_size >= graphics.min_subgroup_size && wave_size <= graphics.max_subgroup_size) {
+		comp_subgroup_size.sType =
+		    vk::StructureType::ePipelineShaderStageRequiredSubgroupSizeCreateInfo;
+		comp_subgroup_size.requiredSubgroupSize = wave_size;
+		comp_shader_stage_info.pNext            = &comp_subgroup_size;
+	}
 
-	vk::DescriptorSetLayout set_layouts[1]  = {};
-	uint32_t                set_layouts_num = 0;
-
-	vk::PushConstantRange push_constant_info[1];
-	uint32_t              push_constant_info_num = 0;
-
-	EXIT_IF(!input_info.stage);
-	CreateLayout(descriptor_cache, set_layouts, set_layouts_num, push_constant_info,
-	             push_constant_info_num, *input_info.stage.program,
-	             vk::ShaderStageFlagBits::eCompute, DescriptorCache::Stage::Compute);
+	std::vector<vk::DescriptorSetLayoutBinding> descriptor_bindings;
+	AddLayoutBindings(descriptor_bindings, *input_info.stage.program,
+	                  vk::ShaderStageFlagBits::eCompute);
+	CreateDescriptorLayout(graphics, pipeline, descriptor_bindings);
+	const vk::PushConstantRange push_constants {vk::ShaderStageFlagBits::eCompute, 0,
+	                                            ShaderRecompiler::IR::NativePushConstantSize};
 
 	vk::PipelineLayoutCreateInfo pipeline_layout_info {};
 	pipeline_layout_info.sType                  = vk::StructureType::ePipelineLayoutCreateInfo;
 	pipeline_layout_info.pNext                  = nullptr;
 	pipeline_layout_info.flags                  = {};
-	pipeline_layout_info.setLayoutCount         = set_layouts_num;
-	pipeline_layout_info.pSetLayouts            = (set_layouts_num > 0 ? set_layouts : nullptr);
-	pipeline_layout_info.pushConstantRangeCount = push_constant_info_num;
-	pipeline_layout_info.pPushConstantRanges =
-	    push_constant_info_num > 0 ? push_constant_info : nullptr;
+	pipeline_layout_info.setLayoutCount         = 1;
+	pipeline_layout_info.pSetLayouts            = &pipeline.descriptor_set_layout;
+	pipeline_layout_info.pushConstantRangeCount = 1;
+	pipeline_layout_info.pPushConstantRanges    = &push_constants;
 
 	EXIT_IF(pipeline.pipeline_layout != nullptr);
 
-	LOGF("PipelineTrace: vkCreatePipelineLayout CS begin set_layouts=%u push_constants=%u\n",
-	     set_layouts_num, push_constant_info_num);
-	result = graphics.device.createPipelineLayout(&pipeline_layout_info, nullptr,
-	                                              &pipeline.pipeline_layout);
+	LOGF("PipelineTrace: vkCreatePipelineLayout CS begin set_layouts=1 push_constants=%u\n",
+	     1u);
+	auto result = graphics.device.createPipelineLayout(&pipeline_layout_info, nullptr,
+	                                                  &pipeline.pipeline_layout);
 	LOGF("PipelineTrace: vkCreatePipelineLayout CS done result=%s layout=%p\n",
 	     VulkanToString(result).c_str(), static_cast<void*>(pipeline.pipeline_layout));
 	EXIT_NOT_IMPLEMENTED(result != vk::Result::eSuccess);
@@ -1123,14 +1025,13 @@ void CreatePipelineInternal(GraphicContext& graphics, DescriptorCache& descripto
 
 	LOGF("PipelineTrace: vkCreateComputePipelines begin layout=%p\n",
 	     static_cast<void*>(pipeline.pipeline_layout));
-	result = graphics.device.createComputePipelines(nullptr, 1, &info, nullptr, &pipeline.pipeline);
+	result = graphics.device.createComputePipelines(driver_cache, 1, &info, nullptr,
+	                                                &pipeline.pipeline);
 	LOGF("PipelineTrace: vkCreateComputePipelines done result=%s pipeline=%p\n",
 	     VulkanToString(result).c_str(), static_cast<void*>(pipeline.pipeline));
 	EXIT_NOT_IMPLEMENTED(result != vk::Result::eSuccess);
 
 	EXIT_NOT_IMPLEMENTED(pipeline.pipeline == nullptr);
-
-	graphics.device.destroyShaderModule(comp_shader_module, nullptr);
 }
 
 } // namespace Libs::Graphics

src/graphics/host_gpu/renderer/render.h

@@ -4,12 +4,14 @@
 #include "common/abi.h"
 #include "common/assert.h"
 #include "common/common.h"
-#include "graphics/host_gpu/renderer/pipeline/descriptorCache.h"
+#include "graphics/host_gpu/renderer/pipeline/descriptors.h"
+#include "graphics/host_gpu/renderer/pipeline/pipelineCache.h"
 #include "graphics/host_gpu/renderer/renderTarget.h"
 #include "graphics/host_gpu/vulkanCommon.h"
 
 #include <array>
-#include <memory>
+#include <optional>
+#include <span>
 #include <vector>
 
 namespace Libs::Graphics {
@@ -23,9 +25,6 @@ class Shader;
 struct GraphicContext;
 struct ShaderBufferResource;
 struct ShaderComputeInputInfo;
-struct CommandSlot;
-struct VulkanBuffer;
-struct VulkanDescriptorSet;
 struct RenderDepthInfo;
 struct RenderColorInfo;
 struct DrawCallInfo;
@@ -49,20 +48,6 @@ enum class CommandBufferDebugOp : uint32_t {
 	Unknown,
 };
 
-class FenceResourceRetainer {
-public:
-	FenceResourceRetainer() = default;
-	~FenceResourceRetainer();
-	KYTY_CLASS_NO_COPY(FenceResourceRetainer);
-
-	void               Retain(std::shared_ptr<void> resource);
-	void               ReleaseAfterFence() noexcept;
-	[[nodiscard]] bool Empty() const noexcept { return m_resources.empty(); }
-
-private:
-	std::vector<std::shared_ptr<void>> m_resources;
-};
-
 struct SubmitInfo {
 	static constexpr uint32_t MaxSemaphores = 3;
 
@@ -91,78 +76,52 @@ struct SubmitInfo {
 
 class CommandBuffer {
 public:
-	explicit CommandBuffer(CommandScheduler& scheduler);
-	~CommandBuffer();
+	~CommandBuffer() = default;
 
 	KYTY_CLASS_NO_COPY(CommandBuffer);
 
 	[[nodiscard]] bool IsInvalid() const;
 
-	void Begin() const;
-	void End() const;
-	void Execute(const SubmitInfo& submit = {});
 	void SetDebugInfo(uint32_t op, uint64_t submit_id, uint32_t arg0 = 0, uint32_t arg1 = 0,
 	                  uint32_t arg2 = 0, uint32_t arg3 = 0, uint64_t arg4 = 0);
 	void BeginRendering(const RenderState& state) const;
 	void EndRendering() const;
-	void WaitForFenceOnly();
-	void WaitForFence();
-	void WaitForFenceAndReset();
-	void RetireBufferAfterFence(std::unique_ptr<VulkanBuffer> buffer);
-	void RetainResourceUntilFence(std::shared_ptr<void> resource);
-	void RecycleDescriptorAfterFence(VulkanDescriptorSet& set);
 
 	[[nodiscard]] vk::CommandBuffer Handle() const;
 	[[nodiscard]] GraphicContext&   GetGraphics() const noexcept { return m_graphics; }
 	[[nodiscard]] RenderContext&    GetContext() const noexcept { return m_context; }
-	[[nodiscard]] bool              IsExecute() const { return m_execute; }
+	[[nodiscard]] HW::Context&      GetRegisters() const noexcept { return *m_registers; }
+	[[nodiscard]] HW::UserConfig&   GetUserConfig() const noexcept { return *m_user_config; }
+	[[nodiscard]] HW::Shader&       GetShaders() const noexcept { return *m_shaders; }
 
 private:
-	void Release();
-	void FinalizeFence(bool reset_recording);
-	void ReleaseResourcesAfterFence();
-	void DeleteBuffersAfterFence();
-	void RecycleDescriptorsAfterFence();
-
-	RenderContext&                             m_context;
-	CommandScheduler&                          m_scheduler;
-	GraphicContext&                            m_graphics;
-	CommandSlot*                               m_slot            = nullptr;
-	bool                                       m_execute         = false;
-	bool                                       m_fence_waited    = false;
-	uint64_t                                   m_submit_seq      = 0;
-	uint32_t                                   m_debug_op        = 0;
-	uint64_t                                   m_debug_submit_id = 0;
-	uint32_t                                   m_debug_arg0      = 0;
-	uint32_t                                   m_debug_arg1      = 0;
-	uint32_t                                   m_debug_arg2      = 0;
-	uint32_t                                   m_debug_arg3      = 0;
-	uint64_t                                   m_debug_arg4      = 0;
-	std::vector<std::unique_ptr<VulkanBuffer>> m_retired_buffers;
-	FenceResourceRetainer                      m_fence_resources;
-	std::vector<VulkanDescriptorSet*>          m_descriptor_sets_after_fence;
-	mutable RenderState                        m_render_state;
-	mutable bool                               m_rendering = false;
-};
-
-class RenderCommandBuffer final: public CommandBuffer {
-public:
-	explicit RenderCommandBuffer(CommandScheduler& scheduler): CommandBuffer(scheduler) {}
-
+	explicit CommandBuffer(CommandScheduler& scheduler);
 	void Bind(HW::Context& registers, HW::UserConfig& user_config, HW::Shader& shaders) noexcept {
 		m_registers   = &registers;
 		m_user_config = &user_config;
 		m_shaders     = &shaders;
 	}
 
-	[[nodiscard]] HW::Context&    GetRegisters() const noexcept { return *m_registers; }
-	[[nodiscard]] HW::UserConfig& GetUserConfig() const noexcept { return *m_user_config; }
-	[[nodiscard]] HW::Shader&     GetShaders() const noexcept { return *m_shaders; }
+	void Begin();
+	void End() const;
 
-private:
-	HW::Context*    m_registers   = nullptr;
-	HW::UserConfig* m_user_config = nullptr;
-	HW::Shader*     m_shaders     = nullptr;
+	RenderContext&      m_context;
+	GraphicContext&     m_graphics;
+	vk::CommandBuffer   m_buffer          = nullptr;
+	uint32_t            m_debug_op        = 0;
+	uint64_t            m_debug_submit_id = 0;
+	uint32_t            m_debug_arg0      = 0;
+	uint32_t            m_debug_arg1      = 0;
+	uint32_t            m_debug_arg2      = 0;
+	uint32_t            m_debug_arg3      = 0;
+	uint64_t            m_debug_arg4      = 0;
+	mutable RenderState m_render_state;
+	mutable bool        m_rendering   = false;
+	HW::Context*        m_registers   = nullptr;
+	HW::UserConfig*     m_user_config = nullptr;
+	HW::Shader*         m_shaders     = nullptr;
+
+	friend class CommandScheduler;
 };
 
 class RenderExecutor {
@@ -170,67 +129,69 @@ public:
 	explicit RenderExecutor(RenderContext& context): m_context(context) {}
 	KYTY_CLASS_NO_COPY(RenderExecutor);
 
-	void DrawIndex(uint64_t submit_id, RenderCommandBuffer& buffer, uint32_t index_type_and_size,
+	void DrawIndex(uint64_t submit_id, CommandBuffer& buffer, uint32_t index_type_and_size,
 	               uint32_t index_count, const void* index_addr, uint32_t flags, uint32_t type,
 	               uint32_t instance_count = 1, uint32_t render_target_slice_offset = 0,
 	               int32_t vertex_offset_add = 0, uint32_t first_instance = 0);
-	void DrawAuto(uint64_t submit_id, RenderCommandBuffer& buffer, uint32_t index_count,
-	              uint32_t flags, uint32_t render_target_slice_offset = 0,
-	              uint32_t instance_count = 1, uint32_t first_vertex = 0,
-	              uint32_t first_instance = 0);
-	void DispatchDirect(uint64_t submit_id, RenderCommandBuffer& buffer, uint32_t thread_group_x,
+	void DrawAuto(uint64_t submit_id, CommandBuffer& buffer, uint32_t index_count, uint32_t flags,
+	              uint32_t render_target_slice_offset = 0, uint32_t instance_count = 1,
+	              uint32_t first_vertex = 0, uint32_t first_instance = 0);
+	void DispatchDirect(uint64_t submit_id, CommandBuffer& buffer, uint32_t thread_group_x,
 	                    uint32_t thread_group_y, uint32_t thread_group_z, uint32_t mode);
 
-	[[nodiscard]] DescriptorCache::PreparedBindings
-	     PrepareBindings(CommandBuffer& buffer, const ShaderStageRuntime& runtime,
-	                     vk::ShaderStageFlags shader_stage, DescriptorCache::Stage stage);
-	void RebindBuffers(CommandBuffer& buffer, DescriptorCache::PreparedBindings& bindings);
-	void RebindImages(CommandBuffer& buffer, DescriptorCache::PreparedBindings& bindings);
+	[[nodiscard]] PreparedBindings PrepareBindings(const ShaderStageRuntime& runtime);
+	void                           FindBuffers(PreparedBindings& bindings);
+	void                           RebindBuffers(PreparedBindings& bindings);
+	void                           RebindImages(PreparedBindings& bindings);
 	void CommitBindings(CommandBuffer& buffer, vk::PipelineBindPoint pipeline_bind_point,
-	                    vk::PipelineLayout layout, DescriptorCache::PreparedBindings& bindings);
+	                    const PipelineCache::Pipeline&     pipeline,
+	                    std::span<PreparedBindings* const> bindings);
 
 private:
 	struct GraphicsBindings {
-		DescriptorCache::PreparedBindings                vertex;
-		std::optional<DescriptorCache::PreparedBindings> pixel;
+		PreparedBindings                vertex;
+		std::optional<PreparedBindings> pixel;
 	};
 
-	[[nodiscard]] DescriptorCache::TextureBinding
-	ResolveTexture(const ShaderRecompiler::IR::ImageResource&   resource,
-	               const ShaderRecompiler::IR::DescriptorValue& value);
-	[[nodiscard]] GraphicsBindings PrepareGraphicsBindings(CommandBuffer&            buffer,
-	                                                       const ShaderStageRuntime& vertex,
+	[[nodiscard]] TextureBinding ResolveTexture(const ShaderRecompiler::IR::ImageResource& resource,
+	                                            const ShaderRecompiler::IR::DescriptorValue& value);
+	[[nodiscard]] GraphicsBindings PrepareGraphicsBindings(const ShaderStageRuntime& vertex,
 	                                                       const ShaderStageRuntime& pixel,
 	                                                       bool                      pixel_active);
-	void ResolveRenderColorTarget(uint64_t submit_id, RenderCommandBuffer& buffer,
+	void ResolveRenderColorTarget(uint64_t submit_id, CommandBuffer& buffer,
 	                              RenderColorInfo& target, uint32_t render_target_slice_offset = 0,
 	                              uint32_t render_target_slot = UINT32_MAX,
 	                              bool ignore_target_mask = false, bool exact_format = false);
-	void ResolveRenderDepthTarget(uint64_t submit_id, RenderCommandBuffer& buffer,
+	void ResolveRenderDepthTarget(uint64_t submit_id, CommandBuffer& buffer,
 	                              RenderDepthInfo& target);
-	[[nodiscard]] bool PrepareDrawRenderState(uint64_t submit_id, RenderCommandBuffer& buffer,
+	[[nodiscard]] bool PrepareDrawRenderState(uint64_t submit_id, CommandBuffer& buffer,
 	                                          const DrawCallInfo& draw,
 	                                          uint32_t            render_target_slice_offset,
 	                                          bool log_setup_phases, DrawRenderState& state);
-	void               ExecutePreparedDraw(uint64_t submit_id, RenderCommandBuffer& buffer,
-	                                       const DrawCallInfo& draw, DrawRenderState& state,
-	                                       vk::PrimitiveTopology topology, const DrawEmitInfo& emit,
-	                                       const DrawIndexBufferSource& index_source,
-	                                       bool primitive_restart_enable, bool log_pipeline_phase,
-	                                       bool set_bind_debug, bool set_auto_debug);
+	void ExecutePreparedDraw(uint64_t submit_id, CommandBuffer& buffer, const DrawCallInfo& draw,
+	                         DrawRenderState& state, vk::PrimitiveTopology topology,
+	                         const DrawEmitInfo& emit, const DrawIndexBufferSource& index_source,
+	                         bool primitive_restart_enable, bool log_pipeline_phase,
+	                         bool set_bind_debug, bool set_auto_debug);
 	[[nodiscard]] RenderState AcquireRenderTargets(CommandBuffer& buffer, RenderColorInfo* colors,
 	                                               uint32_t color_count, RenderDepthInfo& depth);
-	[[nodiscard]] bool        ResolveColorTargets(uint64_t submit_id, RenderCommandBuffer& buffer,
+	[[nodiscard]] bool        ResolveColorTargets(uint64_t submit_id, CommandBuffer& buffer,
 	                                              uint32_t render_target_slice_offset);
 	void                      BindImage(ImageId id, bool storage);
 	void                      BindRenderTarget(ImageId id);
 	void                      TrackImageBinding(ImageId id);
 	void                      ResetBindings();
 	[[nodiscard]] bool        TryConsumeComputeMetaClear(const ShaderComputeInputInfo& input,
-	                                                     const RenderCommandBuffer&    buffer);
+	                                                     const CommandBuffer&          buffer);
 
-	RenderContext&                      m_context;
-	std::vector<std::shared_ptr<Image>> m_bound_images;
+	RenderContext&                        m_context;
+	std::vector<ImageId>                  m_bound_images;
+	std::vector<vk::DescriptorBufferInfo> m_descriptor_buffers;
+	std::vector<vk::DescriptorImageInfo>  m_descriptor_images;
+	std::vector<vk::WriteDescriptorSet>   m_descriptor_writes;
+	std::vector<uint32_t>                 m_image_occurrences;
+	std::array<uint32_t, ShaderRecompiler::IR::NativePushConstantSize / sizeof(uint32_t)>
+	    m_push_constants {};
 
 	friend struct RenderExecutorTestAccess;
 };

src/graphics/host_gpu/renderer/renderCompute.cpp

@@ -11,11 +11,9 @@
 #include "graphics/guest_gpu/hardwareContext.h"
 #include "graphics/host_gpu/graphicContext.h"
 #include "graphics/host_gpu/renderer/image/imageInfo.h"
-#include "graphics/host_gpu/renderer/pipeline/descriptorCache.h"
 #include "graphics/host_gpu/renderer/pipeline/descriptors.h"
 #include "graphics/host_gpu/renderer/pipeline/pipelineCache.h"
 #include "graphics/host_gpu/renderer/pipeline/shaderResourceBarrier.h"
-#include "graphics/host_gpu/renderer/pipeline/shaderSubgroup.h"
 #include "graphics/host_gpu/renderer/render.h"
 #include "graphics/host_gpu/renderer/renderContext.h"
 #include "graphics/host_gpu/vulkanCommon.h"
@@ -46,7 +44,7 @@ static uint64_t BufferDescriptorSize(const ShaderBufferResource& descriptor) {
 }
 
 bool RenderExecutor::TryConsumeComputeMetaClear(const ShaderComputeInputInfo& input,
-                                                const RenderCommandBuffer&    buffer) {
+                                                const CommandBuffer&          buffer) {
 	const auto& program   = *input.stage.program;
 	const auto& resources = *input.stage.resources;
 	if (resources.buffers.size() != program.info.buffers.size()) {
@@ -83,9 +81,8 @@ bool ResolveComputeImageClear(const ShaderComputeInputInfo& input, uint32_t grou
 	const auto& program   = *input.stage.program;
 	const auto& resources = *input.stage.resources;
 	if (program.info.buffers.size() != 1 || resources.buffers.size() != 1 ||
-	    !program.info.images.empty() || !program.info.samplers.empty() ||
-	    !program.info.addresses.empty() || !resources.images.empty() ||
-	    !resources.samplers.empty() || !resources.addresses.empty()) {
+	    !program.info.images.empty() || !program.info.samplers.empty() || program.info.uses_dma ||
+	    !resources.images.empty() || !resources.samplers.empty()) {
 		return false;
 	}
 	const auto& resource   = program.info.buffers.front();
@@ -165,10 +162,11 @@ static bool TryConsumeComputeImageClear(const ShaderComputeInputInfo& input, Com
 	return true;
 }
 
-void RenderExecutor::DispatchDirect(uint64_t submit_id, RenderCommandBuffer& buffer,
+void RenderExecutor::DispatchDirect(uint64_t submit_id, CommandBuffer& buffer,
                                     uint32_t thread_group_x, uint32_t thread_group_y,
                                     uint32_t thread_group_z, uint32_t mode) {
 	EXIT_IF(buffer.IsInvalid());
+	m_context.GetCommandScheduler().PopPendingOperations();
 	auto& ctx    = buffer.GetRegisters();
 	auto& sh_ctx = buffer.GetShaders();
 
@@ -209,16 +207,12 @@ void RenderExecutor::DispatchDirect(uint64_t submit_id, RenderCommandBuffer& buf
 	const auto& cs_regs = sh_ctx.GetCs();
 	const auto& sh_regs = ctx.GetShaderRegisters();
 
-	ShaderComputeInputInfo    input_info {};
-	std::span<const uint32_t> cs_shader;
-	if (!ShaderCompileInfoCS(cs_regs, sh_regs, input_info, cs_shader)) {
-		EXIT("ShaderCompileInfoCS failed for dispatch with CS shader 0x%016" PRIx64 "\n",
-		     cs_regs.cs_regs.data_addr);
-	}
-
+	ShaderComputeInputInfo input_info {};
 	const bool use_thread_dimensions = (mode & DISPATCH_INITIATOR_USE_THREAD_DIMENSIONS) != 0;
+	input_info.dispatch_thread_dimensions = use_thread_dimensions;
+	const auto compute_program =
+	    m_context.GetPipelineCache().GetComputeProgram(cs_regs, sh_regs, input_info);
 	if (use_thread_dimensions) {
-		input_info.dispatch_thread_dimensions = true;
 		input_info.dispatch_threads_num[0]    = thread_group_x;
 		input_info.dispatch_threads_num[1]    = thread_group_y;
 		input_info.dispatch_threads_num[2]    = thread_group_z;
@@ -333,14 +327,19 @@ void RenderExecutor::DispatchDirect(uint64_t submit_id, RenderCommandBuffer& buf
 
 	buffer.EndRendering();
 	auto& pipeline =
-	    m_context.GetPipelineCache().CreateComputePipeline(input_info, sh_ctx.GetCs(), cs_shader);
-	auto bindings = PrepareBindings(buffer, input_info.stage, vk::ShaderStageFlagBits::eCompute,
-	                                DescriptorCache::Stage::Compute);
-	RebindBuffers(buffer, bindings);
-	RebindImages(buffer, bindings);
+	    m_context.GetPipelineCache().CreateComputePipeline(input_info, compute_program);
+	auto bindings = PrepareBindings(input_info.stage);
+	FindBuffers(bindings);
+	if (program.info.uses_dma) {
+		m_context.GetGpuResources().PrepareBda();
+	}
+	RebindBuffers(bindings);
+	RebindImages(bindings);
 
-	auto vk_buffer = buffer.Handle();
-	CommitBindings(buffer, vk::PipelineBindPoint::eCompute, pipeline.pipeline_layout, bindings);
+	auto              vk_buffer        = buffer.Handle();
+	PreparedBindings* descriptor_stage = &bindings;
+	CommitBindings(buffer, vk::PipelineBindPoint::eCompute, pipeline,
+	               std::span {&descriptor_stage, 1u});
 	bool has_storage_writes = HasShaderBufferWrites(input_info.stage);
 	has_storage_writes =
 	    std::any_of(program.info.images.begin(), program.info.images.end(),

src/graphics/host_gpu/renderer/renderContext.cpp

@@ -4,7 +4,6 @@
 #include "common/logging/log.h"
 #include "graphics/guest_gpu/graphicsRun.h"
 #include "graphics/presentation/videoOut.h"
-#include "kernel/pthread.h"
 #include "libs/errno.h"
 
 #include <algorithm>
@@ -13,8 +12,9 @@ namespace Libs::Graphics {
 
 RenderContext::RenderContext(GraphicContext& graphics)
     : m_graphics(graphics), m_render_executor(*this), m_command_scheduler(*this, graphics),
-      m_descriptor_cache(graphics), m_pipeline_cache(graphics, m_descriptor_cache),
-      m_sampler_cache(graphics), m_gpu_resources(graphics, m_command_scheduler) {
+      m_descriptor_heap(graphics, m_command_scheduler.GetMasterSemaphore()),
+      m_pipeline_cache(graphics), m_sampler_cache(graphics),
+      m_gpu_resources(graphics, m_command_scheduler) {
 	EXIT_NOT_IMPLEMENTED(!Common::Thread::IsMainThread());
 }
 
@@ -38,7 +38,7 @@ void RenderContext::ShutdownGpu() {
 	}
 	if (m_video_out != nullptr) {
 		if (m_command_scheduler.Active()) {
-			m_command_scheduler.FinishCurrent();
+			m_command_scheduler.Finish();
 		}
 		m_command_scheduler.DrainPriorityOperations();
 		m_video_out = nullptr;
@@ -55,60 +55,54 @@ VideoOut::VideoOutDriver& RenderContext::GetVideoOut() const {
 	return *m_video_out;
 }
 
-void RenderContext::AddEopEq(LibKernel::EventQueue::KernelEqueue eq, int id) {
-	auto queue = LibKernel::EventQueue::KernelPinEqueue(eq);
-	if (!queue) {
+void RenderContext::AddInterruptEq(LibKernel::EventQueue::KernelEqueue eq, int event_id) {
+	Common::LockGuard lock(m_interrupt_mutex);
+
+	auto it = std::find_if(
+	    m_interrupt_eqs.begin(), m_interrupt_eqs.end(),
+	    [eq, event_id](const auto& entry) { return entry.eq == eq && entry.event_id == event_id; });
+	if (it != m_interrupt_eqs.end()) {
 		return;
 	}
 
-	Common::LockGuard lock(m_eop_mutex);
-
-	auto it = std::find_if(m_eop_eqs.begin(), m_eop_eqs.end(), [eq, id](const auto& entry) {
-		return entry.eq == eq && entry.id == id;
-	});
-	if (it != m_eop_eqs.end()) {
-		return;
-	}
-
-	m_eop_eqs.push_back({eq, std::move(queue), id});
+	m_interrupt_eqs.push_back({eq, event_id});
 }
 
-void RenderContext::DeleteEopEq(LibKernel::EventQueue::KernelEqueue eq, int id) {
-	Common::LockGuard lock(m_eop_mutex);
+void RenderContext::DeleteInterruptEq(LibKernel::EventQueue::KernelEqueue eq, int event_id) {
+	Common::LockGuard lock(m_interrupt_mutex);
 
-	auto it = std::find_if(m_eop_eqs.begin(), m_eop_eqs.end(), [eq, id](const auto& entry) {
-		return entry.eq == eq && entry.id == id;
-	});
-	if (it == m_eop_eqs.end()) {
+	auto it = std::find_if(
+	    m_interrupt_eqs.begin(), m_interrupt_eqs.end(),
+	    [eq, event_id](const auto& entry) { return entry.eq == eq && entry.event_id == event_id; });
+	if (it == m_interrupt_eqs.end()) {
 		return;
 	}
 
-	m_eop_eqs.erase(it);
+	m_interrupt_eqs.erase(it);
 }
 
-void RenderContext::TriggerEopEvent(uint32_t context_id) {
-	std::vector<EopEqRegistration> registrations;
+void RenderContext::TriggerInterrupt(int event_id, uint32_t context_id) {
+	std::vector<InterruptEqRegistration> registrations;
 	{
-		Common::LockGuard lock(m_eop_mutex);
-		registrations = m_eop_eqs;
+		Common::LockGuard lock(m_interrupt_mutex);
+		for (const auto& registration: m_interrupt_eqs) {
+			if (registration.event_id == event_id) {
+				registrations.push_back(registration);
+			}
+		}
 	}
 
 	for (const auto& registration: registrations) {
 		const auto result = LibKernel::EventQueue::KernelTriggerEvent(
-		    registration.eq, static_cast<uintptr_t>(registration.id),
+		    registration.eq, static_cast<uintptr_t>(registration.event_id),
 		    LibKernel::EventQueue::KERNEL_EVFILT_GRAPHICS,
 		    reinterpret_cast<void*>(static_cast<uintptr_t>(context_id)));
 		if (result == LibKernel::KERNEL_ERROR_EBADF || result == LibKernel::KERNEL_ERROR_ENOENT) {
-			DeleteEopEq(registration.eq, registration.id);
+			DeleteInterruptEq(registration.eq, registration.event_id);
 			continue;
 		}
 		EXIT_NOT_IMPLEMENTED(result != OK);
 	}
-
-	auto tsc    = LibKernel::KernelReadTsc();
-	auto result = LibKernel::EventQueue::KernelTriggerUserEventForAll(AGC_USER_INTERRUPT_EVENT,
-	                                                                  reinterpret_cast<void*>(tsc));
-	EXIT_NOT_IMPLEMENTED(result != OK && result != LibKernel::KERNEL_ERROR_ENOENT);
 }
 
 } // namespace Libs::Graphics

src/graphics/host_gpu/renderer/renderContext.h

@@ -10,7 +10,7 @@
 #include "graphics/host_gpu/renderer/cache/samplerCache.h"
 #include "graphics/host_gpu/renderer/cache/textureCache.h"
 #include "graphics/host_gpu/renderer/commandScheduler.h"
-#include "graphics/host_gpu/renderer/pipeline/descriptorCache.h"
+#include "graphics/host_gpu/renderer/pipeline/descriptorHeap.h"
 #include "graphics/host_gpu/renderer/pipeline/pipelineCache.h"
 #include "kernel/eventQueue.h"
 
@@ -23,7 +23,6 @@ class VideoOutDriver;
 
 namespace Libs::Graphics {
 
-constexpr int AGC_USER_INTERRUPT_EVENT = 0x1800;
 class GuestGpu;
 
 class RenderContext {
@@ -41,37 +40,36 @@ public:
 	Common::Mutex&      GetMutex() { return m_mutex; }
 	CommandScheduler&   GetCommandScheduler() { return m_command_scheduler; }
 	PipelineCache&      GetPipelineCache() { return m_pipeline_cache; }
-	DescriptorCache&    GetDescriptorCache() { return m_descriptor_cache; }
+	DescriptorHeap&     GetDescriptorHeap() { return m_descriptor_heap; }
 	SamplerCache&       GetSamplerCache() { return m_sampler_cache; }
 	GpuResourceManager& GetGpuResources() { return m_gpu_resources; }
 	BufferCache&        GetBufferCache() { return m_gpu_resources.GetBufferCache(); }
 	TextureCache&       GetTextureCache() { return m_gpu_resources.GetTextureCache(); }
 	RenderExecutor&     GetRenderExecutor() { return m_render_executor; }
 
-	void AddEopEq(LibKernel::EventQueue::KernelEqueue eq, int id);
-	void DeleteEopEq(LibKernel::EventQueue::KernelEqueue eq, int id);
-	void TriggerEopEvent(uint32_t context_id);
+	void AddInterruptEq(LibKernel::EventQueue::KernelEqueue eq, int event_id);
+	void DeleteInterruptEq(LibKernel::EventQueue::KernelEqueue eq, int event_id);
+	void TriggerInterrupt(int event_id, uint32_t context_id);
 
 private:
-	struct EopEqRegistration {
-		LibKernel::EventQueue::KernelEqueue    eq = LibKernel::EventQueue::KERNEL_EQUEUE_INVALID;
-		LibKernel::EventQueue::KernelEqueueRef queue;
-		int                                    id = 0;
+	struct InterruptEqRegistration {
+		LibKernel::EventQueue::KernelEqueue eq       = LibKernel::EventQueue::KERNEL_EQUEUE_INVALID;
+		int                                 event_id = 0;
 	};
 
 	GraphicContext&           m_graphics;
 	Common::Mutex             m_mutex;
 	RenderExecutor            m_render_executor;
 	CommandScheduler          m_command_scheduler;
-	DescriptorCache           m_descriptor_cache;
+	DescriptorHeap            m_descriptor_heap;
 	PipelineCache             m_pipeline_cache;
 	SamplerCache              m_sampler_cache;
 	GpuResourceManager        m_gpu_resources;
 	std::unique_ptr<GuestGpu> m_gpu;
 	VideoOut::VideoOutDriver* m_video_out = nullptr;
 
-	Common::Mutex                  m_eop_mutex;
-	std::vector<EopEqRegistration> m_eop_eqs;
+	Common::Mutex                        m_interrupt_mutex;
+	std::vector<InterruptEqRegistration> m_interrupt_eqs;
 };
 
 } // namespace Libs::Graphics

src/graphics/host_gpu/renderer/renderDraw.cpp

@@ -15,10 +15,8 @@
 #include "graphics/host_gpu/renderer/colorRenderTarget.h"
 #include "graphics/host_gpu/renderer/debug.h"
 #include "graphics/host_gpu/renderer/depthRenderTarget.h"
-#include "graphics/host_gpu/renderer/pipeline/descriptorCache.h"
 #include "graphics/host_gpu/renderer/pipeline/pipelineCache.h"
 #include "graphics/host_gpu/renderer/pipeline/shaderResourceBarrier.h"
-#include "graphics/host_gpu/renderer/pipeline/shaderSubgroup.h"
 #include "graphics/host_gpu/renderer/render.h"
 #include "graphics/host_gpu/renderer/renderContext.h"
 #include "graphics/host_gpu/vulkanCommon.h"
@@ -105,7 +103,7 @@ static bool IsDualSourceBlendFactor(uint32_t factor) {
 }
 
 static void LogFramebufferSkip(const char* draw_name, const RenderColorInfo& color,
-                               const RenderDepthInfo& depth, const RenderCommandBuffer& buffer,
+                               const RenderDepthInfo& depth, const CommandBuffer& buffer,
                                uint32_t index_count, uint32_t flags) {
 	const auto& ctx  = buffer.GetRegisters();
 	const auto& ucfg = buffer.GetUserConfig();
@@ -128,7 +126,7 @@ static void LogFramebufferSkip(const char* draw_name, const RenderColorInfo& col
 	    static_cast<uint32_t>(ucfg.GetPrimType()), index_count, flags);
 }
 
-static void LogMrtState(const char* draw_name, const RenderCommandBuffer& buffer,
+static void LogMrtState(const char* draw_name, const CommandBuffer& buffer,
                         const ShaderPixelInputInfo& ps_input_info) {
 	const auto& ctx            = buffer.GetRegisters();
 	const auto& sh_regs        = ctx.GetShaderRegisters();
@@ -142,18 +140,6 @@ static void LogMrtState(const char* draw_name, const RenderCommandBuffer& buffer
 	                   (bc0.separate_alpha_blend && (IsDualSourceBlendFactor(bc0.alpha_srcblend) ||
 	                                                 IsDualSourceBlendFactor(bc0.alpha_destblend)));
 
-	for (uint32_t i = 1; i < 8; i++) {
-		const auto& rt = ctx.GetRenderTarget(i);
-		if (rt.base.addr != 0 || ps_input_info.target_output_mode[i] != 0 ||
-		    ((rt_mask >> (i * 4u)) & 0x0fu) != 0 || ((cb_shader_mask >> (i * 4u)) & 0x0fu) != 0) {
-			interesting = true;
-		}
-	}
-
-	if (!interesting) {
-		return;
-	}
-
 	auto log_id = g_mrt_state_log_count.fetch_add(1);
 	if (log_id >= 32) {
 		return;
@@ -191,7 +177,7 @@ static void LogMrtState(const char* draw_name, const RenderCommandBuffer& buffer
 }
 
 static void LogDrawTargetState(const char* draw_name, const RenderColorInfo& color,
-                               const RenderDepthInfo& depth, const RenderCommandBuffer& buffer,
+                               const RenderDepthInfo& depth, const CommandBuffer& buffer,
                                const ShaderPixelInputInfo& ps_input_info, uint32_t index_count,
                                uint32_t flags) {
 	const auto& ctx  = buffer.GetRegisters();
@@ -245,7 +231,7 @@ static void LogDrawTargetState(const char* draw_name, const RenderColorInfo& col
 	LogMrtState(draw_name, buffer, ps_input_info);
 }
 
-static void LogDrawInputState(const RenderCommandBuffer& buffer, const RenderColorInfo& color,
+static void LogDrawInputState(const CommandBuffer& buffer, const RenderColorInfo& color,
                               const ShaderVertexInputInfo& vs_input_info,
                               uint32_t index_type_and_size, uint32_t index_count,
                               const void* index_addr) {
@@ -323,17 +309,14 @@ static void LogDrawInputState(const RenderCommandBuffer& buffer, const RenderCol
 	}
 }
 
-static PipelineDynamicParameters BuildGraphicsDynamicParams(const RenderCommandBuffer& buffer,
-                                                            const RenderColorInfo*     colors,
-                                                            uint32_t                   color_count,
-                                                            const RenderDepthInfo&     depth) {
+static void SetGraphicsDynamicParams(const CommandBuffer& buffer, vk::CommandBuffer vk_buffer,
+                                     const RenderColorInfo* colors, uint32_t color_count,
+                                     const RenderDepthInfo& depth) {
+	KYTY_PROFILER_FUNCTION();
+
 	EXIT_IF(colors == nullptr);
 	const auto& ctx = buffer.GetRegisters();
 
-	PipelineDynamicParameters ret {};
-	ret.color_write_count   = color_count;
-	ret.stencil_test_enable = depth.stencil_test_enable;
-
 	const auto&  vp = ctx.GetScreenViewport();
 	vk::Extent2D framebuffer_extent {};
 	if (color_count > 0 && colors[0].image_id) {
@@ -347,68 +330,22 @@ static PipelineDynamicParameters BuildGraphicsDynamicParams(const RenderCommandB
 
 	const auto final_scissor = calc_final_scissor(vp, ctx.GetScanModeControl(), framebuffer_extent);
 
-	ret.viewport_scale[0]  = vp.viewports[0].xscale;
-	ret.viewport_scale[1]  = vp.viewports[0].yscale;
-	ret.viewport_scale[2]  = vp.viewports[0].zscale;
-	ret.viewport_offset[0] = vp.viewports[0].xoffset;
-	ret.viewport_offset[1] = vp.viewports[0].yoffset;
-	ret.viewport_offset[2] = vp.viewports[0].zoffset;
-	ret.scissor_ltrb[0]    = final_scissor.left;
-	ret.scissor_ltrb[1]    = final_scissor.top;
-	ret.scissor_ltrb[2]    = final_scissor.right;
-	ret.scissor_ltrb[3]    = final_scissor.bottom;
-	ret.line_width         = ctx.GetLineWidth();
-	ret.stencil_front      = depth.stencil_dynamic_front;
-	ret.stencil_back       = depth.stencil_dynamic_back;
-
-	const auto& mode        = ctx.GetModeControl();
-	const auto& poly_offset = ctx.GetPolyOffset();
-	const bool  use_front   = mode.poly_offset_front_enable && !mode.cull_front;
-	const bool  use_back    = mode.poly_offset_back_enable && !mode.cull_back;
-	ret.depth_bias_enable   = use_front || use_back;
-	if (ret.depth_bias_enable) {
-		// Vulkan has one bias for both faces. Prefer a visible front face when both are enabled.
-		const float guest_constant_factor =
-		    use_front ? poly_offset.front_offset : poly_offset.back_offset;
-		ret.depth_bias_constant_factor =
-		    ConvertPolygonOffsetConstantFactor(guest_constant_factor, poly_offset, depth.format);
-		ret.depth_bias_clamp = poly_offset.clamp;
-		ret.depth_bias_slope_factor =
-		    (use_front ? poly_offset.front_scale : poly_offset.back_scale) / 16.0f;
-	}
-
-	// Color-control operation selects special color-buffer paths, not the normal component write
-	// mask. Attachment availability therefore follows the target write mask.
-	for (uint32_t i = 0; i < RENDER_COLOR_ATTACHMENTS_MAX; i++) {
-		ret.color_write_enable[i] =
-		    (i < color_count &&
-		     render_target_mask_slot(ctx.GetRenderTargetMask(), colors[i].target_slot) != 0);
-	}
-
-	return ret;
-}
-
-static void SetDynamicParams(const RenderCommandBuffer& buffer, vk::CommandBuffer vk_buffer,
-                             const PipelineDynamicParameters& dynamic_params) {
-	KYTY_PROFILER_FUNCTION();
-
 	vk::Viewport viewport {};
-	viewport.x        = dynamic_params.viewport_offset[0] - dynamic_params.viewport_scale[0];
-	viewport.y        = dynamic_params.viewport_offset[1] - dynamic_params.viewport_scale[1];
-	viewport.width    = dynamic_params.viewport_scale[0] * 2.0f;
-	viewport.height   = dynamic_params.viewport_scale[1] * 2.0f;
-	viewport.minDepth = dynamic_params.viewport_offset[2];
-	viewport.maxDepth = dynamic_params.viewport_scale[2] + dynamic_params.viewport_offset[2];
+	viewport.x        = vp.viewports[0].xoffset - vp.viewports[0].xscale;
+	viewport.y        = vp.viewports[0].yoffset - vp.viewports[0].yscale;
+	viewport.width    = vp.viewports[0].xscale * 2.0f;
+	viewport.height   = vp.viewports[0].yscale * 2.0f;
+	viewport.minDepth = vp.viewports[0].zoffset;
+	viewport.maxDepth = vp.viewports[0].zscale + vp.viewports[0].zoffset;
 	vk_buffer.setViewport(0, 1, &viewport);
 
 	vk::Rect2D scissor {};
-	scissor.offset = {dynamic_params.scissor_ltrb[0], dynamic_params.scissor_ltrb[1]};
-	scissor.extent = {
-	    static_cast<uint32_t>(dynamic_params.scissor_ltrb[2] - dynamic_params.scissor_ltrb[0]),
-	    static_cast<uint32_t>(dynamic_params.scissor_ltrb[3] - dynamic_params.scissor_ltrb[1])};
+	scissor.offset = {final_scissor.left, final_scissor.top};
+	scissor.extent = {static_cast<uint32_t>(final_scissor.right - final_scissor.left),
+	                  static_cast<uint32_t>(final_scissor.bottom - final_scissor.top)};
 	vk_buffer.setScissor(0, 1, &scissor);
 
-	float line_width = dynamic_params.line_width;
+	float line_width = ctx.GetLineWidth();
 	if (line_width != 1.0f) {
 		static bool logged = false;
 		if (!logged) {
@@ -421,26 +358,36 @@ static void SetDynamicParams(const RenderCommandBuffer& buffer, vk::CommandBuffe
 	}
 	vk_buffer.setLineWidth(line_width);
 
-	vk_buffer.setDepthBiasEnable(dynamic_params.depth_bias_enable ? VK_TRUE : VK_FALSE);
-	if (dynamic_params.depth_bias_enable) {
-		vk_buffer.setDepthBias(dynamic_params.depth_bias_constant_factor,
-		                       dynamic_params.depth_bias_clamp,
-		                       dynamic_params.depth_bias_slope_factor);
+	const auto& mode              = ctx.GetModeControl();
+	const auto& poly_offset       = ctx.GetPolyOffset();
+	const bool  use_front         = mode.poly_offset_front_enable && !mode.cull_front;
+	const bool  use_back          = mode.poly_offset_back_enable && !mode.cull_back;
+	const bool  depth_bias_enable = use_front || use_back;
+	vk_buffer.setDepthBiasEnable(depth_bias_enable ? VK_TRUE : VK_FALSE);
+	if (depth_bias_enable) {
+		// Vulkan has one bias for both faces. Prefer a visible front face when both are enabled.
+		const float guest_constant_factor =
+		    use_front ? poly_offset.front_offset : poly_offset.back_offset;
+		const float constant_factor =
+		    ConvertPolygonOffsetConstantFactor(guest_constant_factor, poly_offset, depth.format);
+		const float slope_factor =
+		    (use_front ? poly_offset.front_scale : poly_offset.back_scale) / 16.0f;
+		vk_buffer.setDepthBias(constant_factor, poly_offset.clamp, slope_factor);
 	}
 
-	if (dynamic_params.stencil_test_enable) {
+	if (depth.stencil_test_enable) {
 		vk_buffer.setStencilCompareMask(vk::StencilFaceFlagBits::eFront,
-		                                dynamic_params.stencil_front.compareMask);
+		                                depth.stencil_dynamic_front.compareMask);
 		vk_buffer.setStencilCompareMask(vk::StencilFaceFlagBits::eBack,
-		                                dynamic_params.stencil_back.compareMask);
+		                                depth.stencil_dynamic_back.compareMask);
 		vk_buffer.setStencilWriteMask(vk::StencilFaceFlagBits::eFront,
-		                              dynamic_params.stencil_front.writeMask);
+		                              depth.stencil_dynamic_front.writeMask);
 		vk_buffer.setStencilWriteMask(vk::StencilFaceFlagBits::eBack,
-		                              dynamic_params.stencil_back.writeMask);
+		                              depth.stencil_dynamic_back.writeMask);
 		vk_buffer.setStencilReference(vk::StencilFaceFlagBits::eFront,
-		                              dynamic_params.stencil_front.reference);
+		                              depth.stencil_dynamic_front.reference);
 		vk_buffer.setStencilReference(vk::StencilFaceFlagBits::eBack,
-		                              dynamic_params.stencil_back.reference);
+		                              depth.stencil_dynamic_back.reference);
 	}
 
 #if defined(__APPLE__)
@@ -448,11 +395,15 @@ static void SetDynamicParams(const RenderCommandBuffer& buffer, vk::CommandBuffe
 	// eColorWriteEnableEXT dynamic state and relies on the static colorWriteMask instead.
 #else
 	vk::Bool32 enable[RENDER_COLOR_ATTACHMENTS_MAX] = {};
-	for (uint32_t i = 0; i < dynamic_params.color_write_count; i++) {
-		enable[i] = (dynamic_params.color_write_enable[i] ? VK_TRUE : VK_FALSE);
+	// Color-control operation selects special color-buffer paths, not the normal component write
+	// mask. Attachment availability therefore follows the target write mask.
+	for (uint32_t i = 0; i < color_count; i++) {
+		enable[i] = render_target_mask_slot(ctx.GetRenderTargetMask(), colors[i].target_slot) != 0
+		                ? VK_TRUE
+		                : VK_FALSE;
 	}
-	if (dynamic_params.color_write_count != 0) {
-		vk_buffer.setColorWriteEnableEXT(dynamic_params.color_write_count, enable);
+	if (color_count != 0) {
+		vk_buffer.setColorWriteEnableEXT(color_count, enable);
 	}
 #endif
 }
@@ -460,7 +411,7 @@ static void SetDynamicParams(const RenderCommandBuffer& buffer, vk::CommandBuffe
 static bool DrawHasValidVertexShader(const HW::Shader& sh_ctx) {
 
 	const auto& vs = sh_ctx.GetVs();
-	return vs.gs_regs.chksum != 0 && ShaderAddressValid(vs.es_regs.data_addr);
+	return ShaderAddressValid(vs.es_regs.data_addr);
 }
 
 static bool PixelShaderHasDepthOrCoverageSideEffects(const HW::ShaderRegisters& sh_regs) {
@@ -470,7 +421,7 @@ static bool PixelShaderHasDepthOrCoverageSideEffects(const HW::ShaderRegisters&
 	       db.shader_execute_on_noop;
 }
 
-static bool ShouldSkipGeShader(const RenderCommandBuffer& buffer) {
+static bool ShouldSkipGeShader(const CommandBuffer& buffer) {
 	const auto& ctx         = buffer.GetRegisters();
 	const auto& ucfg        = buffer.GetUserConfig();
 	const auto& sh_ctx      = buffer.GetShaders();
@@ -492,22 +443,20 @@ static bool ShouldSkipGeShader(const RenderCommandBuffer& buffer) {
 	};
 
 	const bool ps5_ngg_vertex_path = stages == 0x02002000 && vertex_info.es_regs.data_addr != 0 &&
-	                                 vertex_info.gs_regs.chksum != 0 &&
 	                                 sh_regs.m_vgtGsMaxVertOut == 0x00000000 &&
 	                                 is_known_gs_out_prim_type(sh_regs.m_vgtGsOutPrimType);
 
 	const bool unsupported_stage_mask = (stages != 0 && stages != 0x02002000);
 	const bool unsupported_gs_stage = (vertex_info.es_regs.data_addr != 0 &&
 	                                   vertex_info.gs_regs.data_addr != 0 && !ps5_ngg_vertex_path);
-	const bool ge_group_size =
-	    ge_cntl.primitive_group_size > 0x0040 || ge_cntl.vertex_group_size > 0x0040;
+	// GE_CNTL group sizes control guest scheduling and do not constrain the host vertex path.
 	const bool ge_shader_regs =
 	    (sh_regs.m_geNggSubgrpCntl != 0x00000000 && sh_regs.m_geNggSubgrpCntl != 0x00000001) ||
 	    sh_regs.m_vgtGsMaxVertOut != 0x00000000 ||
 	    !is_known_gs_out_prim_type(sh_regs.m_vgtGsOutPrimType) ||
 	    sh_regs.m_geMaxOutputPerSubgroup > 0x00000040;
 
-	if (unsupported_stage_mask || unsupported_gs_stage || ge_group_size || ge_shader_regs) {
+	if (unsupported_stage_mask || unsupported_gs_stage || ge_shader_regs) {
 		static std::once_flag warning_once;
 		std::call_once(warning_once, [] {
 			std::printf("Warning: game uses unsupported graphics pipelines; some draw calls were "
@@ -532,15 +481,15 @@ static bool ShouldSkipGeShader(const RenderCommandBuffer& buffer) {
 }
 
 struct DrawRenderState {
-	RenderDepthInfo           depth_info;
-	RenderColorInfo           color_info[RENDER_COLOR_ATTACHMENTS_MAX] = {};
-	uint32_t                  color_count                              = 0;
-	bool                      ps_active                                = true;
-	RenderState               rendering;
-	ShaderVertexInputInfo     vs_input_info;
-	ShaderPixelInputInfo      ps_input_info;
-	std::span<const uint32_t> vs_shader;
-	std::span<const uint32_t> ps_shader;
+	RenderDepthInfo       depth_info;
+	RenderColorInfo       color_info[RENDER_COLOR_ATTACHMENTS_MAX] = {};
+	uint32_t              color_count                              = 0;
+	bool                  ps_active                                = true;
+	RenderState           rendering;
+	ShaderVertexInputInfo vs_input_info;
+	ShaderPixelInputInfo  ps_input_info;
+	ShaderProgram         vertex_program;
+	ShaderProgram         pixel_program;
 };
 
 struct DrawCallInfo {
@@ -553,8 +502,7 @@ struct DrawCallInfo {
 };
 
 static bool ResolveDccAttachmentClear(TextureCache& cache, const RenderColorInfo& target,
-                                      const ImageViewInfo& view,
-                                      vk::ClearColorValue& clear_value) {
+                                      const ImageViewInfo& view, vk::ClearColorValue& clear_value) {
 	if (target.desc.info.metadata.kind != ImageMetadataKind::Dcc) {
 		return false;
 	}
@@ -630,7 +578,7 @@ RenderState RenderExecutor::AcquireRenderTargets(CommandBuffer& buffer, RenderCo
 	for (uint32_t i = 0; i < color_count; i++) {
 		auto& target = colors[i];
 		EXIT_IF(!target.image_id);
-		const auto old_image = cache.ResolveOwner(target.image_id);
+		const auto old_image = cache.m_slot_images.try_get(target.image_id);
 		if (old_image == nullptr || (!old_image->registered && !old_image->info.data.Empty()) ||
 		    old_image->binding.needs_rebind) {
 			if (old_image != nullptr) {
@@ -641,6 +589,15 @@ RenderState RenderExecutor::AcquireRenderTargets(CommandBuffer& buffer, RenderCo
 		}
 		target.image_view = cache.FindRenderTarget(target.image_id, target.desc);
 		auto& image       = cache.GetImage(target.image_id);
+		SetVulkanObjectNameF(m_context.GetGraphics().device, image.backing.image,
+		                     "Kyty.MRT{}.Image[guest=0x{:016x} size=0x{:x} format={}]",
+		                     target.target_slot, image.info.data.address, image.info.data.size,
+		                     static_cast<uint32_t>(image.info.pixel_format));
+		SetVulkanObjectNameF(m_context.GetGraphics().device, target.image_view,
+		                     "Kyty.MRT{}.View[guest=0x{:016x} mip={} layer={}+{}]",
+		                     target.target_slot, image.info.data.address,
+		                     target.desc.view_info.base_level, target.desc.view_info.base_layer,
+		                     target.desc.view_info.layer_count);
 		EXIT_IF(image.backing.samples != target.samples || target.image_view == nullptr);
 		if (attachment_samples == 0) {
 			attachment_samples = target.samples;
@@ -657,15 +614,15 @@ RenderState RenderExecutor::AcquireRenderTargets(CommandBuffer& buffer, RenderCo
 		              ImageSubresourceRange {view.base_level, view.level_count, view.base_layer,
 		                                     view.layer_count},
 		              buffer.Handle());
-		state.width                 = std::min(state.width, target.extent.width);
-		state.height                = std::min(state.height, target.extent.height);
-		state.num_layers            = std::min(state.num_layers, view.layer_count);
-		auto& attachment            = state.color_attachments[i];
-		attachment.image_view       = target.image_view;
-		attachment.image_layout     = layout;
-		attachment.clear_value      = target.color_clear_value.uint32;
+		state.width             = std::min(state.width, target.extent.width);
+		state.height            = std::min(state.height, target.extent.height);
+		state.num_layers        = std::min(state.num_layers, view.layer_count);
+		auto& attachment        = state.color_attachments[i];
+		attachment.image_view   = target.image_view;
+		attachment.image_layout = layout;
+		attachment.clear_value  = target.color_clear_value.uint32;
 		vk::ClearColorValue metadata_clear_value {};
-		const bool metadata_clear =
+		const bool          metadata_clear =
 		    ResolveDccAttachmentClear(cache, target, view, metadata_clear_value);
 		if (metadata_clear) {
 			attachment.clear_value = metadata_clear_value.uint32;
@@ -673,7 +630,7 @@ RenderState RenderExecutor::AcquireRenderTargets(CommandBuffer& buffer, RenderCo
 		attachment.is_clear = target.color_clear_enable || metadata_clear;
 	}
 	if (depth.image_id) {
-		const auto owner = cache.ResolveOwner(depth.image_id);
+		const auto owner = cache.m_slot_images.try_get(depth.image_id);
 		if (owner == nullptr || !owner->registered || owner->binding.needs_rebind) {
 			EXIT("depth target changed after render-state discovery\n");
 		}
@@ -690,6 +647,14 @@ RenderState RenderExecutor::AcquireRenderTargets(CommandBuffer& buffer, RenderCo
 			EXIT("failed to consume HTile clear state\n");
 		}
 		auto& image = cache.GetImage(depth.image_id);
+		SetVulkanObjectNameF(m_context.GetGraphics().device, image.backing.image,
+		                     "Kyty.DepthTarget.Image[guest=0x{:016x} size=0x{:x} format={}]",
+		                     image.info.data.address, image.info.data.size,
+		                     static_cast<uint32_t>(image.info.pixel_format));
+		SetVulkanObjectNameF(m_context.GetGraphics().device, depth.image_view,
+		                     "Kyty.DepthTarget.View[guest=0x{:016x} layer={}+{}]",
+		                     image.info.data.address, depth.desc.view_info.base_layer,
+		                     depth.desc.view_info.layer_count);
 		EXIT_IF(depth.image_view == nullptr || image.backing.samples != depth.samples);
 		if (attachment_samples == 0) {
 			attachment_samples = depth.samples;
@@ -739,11 +704,12 @@ RenderState RenderExecutor::AcquireRenderTargets(CommandBuffer& buffer, RenderCo
 	return state;
 }
 
-static bool DrawHasActivePixelShader(const RenderCommandBuffer& buffer) {
+static bool DrawHasActivePixelShader(const CommandBuffer& buffer) {
 	const auto& ctx              = buffer.GetRegisters();
 	const auto& sh_regs          = ctx.GetShaderRegisters();
 	const bool  has_color_output = (ctx.GetRenderTargetMask() & sh_regs.m_cbShaderMask) != 0;
-	return has_color_output || PixelShaderHasDepthOrCoverageSideEffects(sh_regs);
+	return ShaderAddressValid(buffer.GetShaders().GetPs().ps_regs.data_addr) &&
+	       (has_color_output || PixelShaderHasDepthOrCoverageSideEffects(sh_regs));
 }
 
 enum class CbColorMode : uint8_t {
@@ -755,7 +721,7 @@ enum class CbColorMode : uint8_t {
 	DccDecompress      = 6,
 };
 
-static bool ConsumeMetadataColorOperation(const RenderCommandBuffer& buffer) {
+static bool ConsumeMetadataColorOperation(const CommandBuffer& buffer) {
 	const auto& ctx  = buffer.GetRegisters();
 	const auto  mode = ctx.GetColorControl().mode;
 	// These special modes run color-buffer metadata or decompression operations. The shader is a
@@ -783,11 +749,10 @@ struct DrawIndexBufferSource {
 };
 
 struct PreparedIndexBuffer {
-	std::shared_ptr<void> owner;
-	vk::Buffer            buffer = nullptr;
-	uint64_t              size   = 0;
-	vk::DeviceSize        offset = 0;
-	vk::IndexType         type   = vk::IndexType::eUint16;
+	vk::Buffer     buffer = nullptr;
+	uint64_t       size   = 0;
+	vk::DeviceSize offset = 0;
+	vk::IndexType  type   = vk::IndexType::eUint16;
 };
 
 static uint64_t VertexBufferDescriptorSize(const ShaderVertexInputBuffer& buffer) {
@@ -796,10 +761,10 @@ static uint64_t VertexBufferDescriptorSize(const ShaderVertexInputBuffer& buffer
 }
 
 struct VertexBufferRange {
-	uint64_t      base_address  = 0;
-	uint64_t      requested_end = 0;
-	uint64_t      acquired_end  = 0;
-	BufferBinding binding;
+	uint64_t                     base_address  = 0;
+	uint64_t                     requested_end = 0;
+	uint64_t                     acquired_end  = 0;
+	std::pair<Buffer*, uint64_t> binding;
 
 	[[nodiscard]] uint64_t RequestedSize() const { return requested_end - base_address; }
 };
@@ -807,14 +772,12 @@ struct VertexBufferRange {
 struct PreparedVertexBuffers {
 	static constexpr uint32_t MaxBuffers = ShaderVertexInputInfo::RES_MAX;
 
-	std::array<vk::Buffer, MaxBuffers>            buffers {};
-	std::array<vk::DeviceSize, MaxBuffers>        offsets {};
-	std::array<std::shared_ptr<void>, MaxBuffers> owners {};
-	uint32_t                                      count       = 0;
-	uint32_t                                      owner_count = 0;
+	std::array<vk::Buffer, MaxBuffers>     buffers {};
+	std::array<vk::DeviceSize, MaxBuffers> offsets {};
+	uint32_t                               count = 0;
 };
 
-static PreparedVertexBuffers AcquireVertexBuffers(RenderCommandBuffer&         buffer,
+static PreparedVertexBuffers AcquireVertexBuffers(CommandBuffer&               buffer,
                                                   const ShaderVertexInputInfo& vs_input_info) {
 	EXIT_IF(vs_input_info.buffers_num < 0 ||
 	        vs_input_info.buffers_num > ShaderVertexInputInfo::RES_MAX);
@@ -861,21 +824,22 @@ static PreparedVertexBuffers AcquireVertexBuffers(RenderCommandBuffer&         b
 		const auto size =
 		    Libs::LibKernel::Memory::ClampRangeSize(range.base_address, range.RequestedSize());
 		range.acquired_end = range.base_address + size;
-		range.binding      = cache.ObtainBuffer(buffer, range.base_address, size);
+		range.binding      = cache.ObtainBuffer(range.base_address, size, false);
+		SetVulkanObjectNameF(
+		    buffer.GetContext().GetGraphics().device, range.binding.first->Handle(),
+		    "Kyty.VertexBufferRange[guest=0x{:016x} size=0x{:x}]", range.base_address, size);
 	}
 
 	// Rebuild slot bindings, offsetting non-empty slots into their acquired merged range.
 	PreparedVertexBuffers prepared;
-	prepared.count = static_cast<uint32_t>(vs_input_info.buffers_num);
-	std::shared_ptr<Buffer> null_owner;
-	vk::Buffer              null_buffer = nullptr;
+	prepared.count         = static_cast<uint32_t>(vs_input_info.buffers_num);
+	vk::Buffer null_buffer = nullptr;
 	for (int i = 0; i < vs_input_info.buffers_num; i++) {
 		const auto& vertex = vs_input_info.buffers[i];
 		const auto  size   = VertexBufferDescriptorSize(vertex);
 		if (size == 0) {
-			if (null_owner == nullptr) {
-				null_owner  = cache.ObtainNullBuffer();
-				null_buffer = null_owner->Handle();
+			if (null_buffer == nullptr) {
+				null_buffer = cache.GetBuffer(NULL_BUFFER_ID).Handle();
 			}
 			prepared.buffers[i] = null_buffer;
 			prepared.offsets[i] = 0;
@@ -892,24 +856,19 @@ static PreparedVertexBuffers AcquireVertexBuffers(RenderCommandBuffer&         b
 			     vertex.addr);
 		}
 
-		prepared.buffers[i] = range->binding.buffer;
-		prepared.offsets[i] = range->binding.offset + vertex.addr - range->base_address;
+		prepared.buffers[i] = range->binding.first->Handle();
+		prepared.offsets[i] = range->binding.second + vertex.addr - range->base_address;
+		SetVulkanObjectNameF(
+		    buffer.GetContext().GetGraphics().device, prepared.buffers[i],
+		    "Kyty.VertexBuffer[slot={} guest=0x{:016x} size=0x{:x} stride={} records={}]", i,
+		    vertex.addr, size, vertex.stride, vertex.num_records);
 	}
 
-	if (null_owner != nullptr) {
-		prepared.owners[prepared.owner_count++] = std::move(null_owner);
-	}
-	for (uint32_t i = 0; i < merged_count; i++) {
-		if (merged_ranges[i].binding.owner != nullptr) {
-			EXIT_IF(prepared.owner_count >= PreparedVertexBuffers::MaxBuffers);
-			prepared.owners[prepared.owner_count++] = std::move(merged_ranges[i].binding.owner);
-		}
-	}
 	return prepared;
 }
 
-static void SetDrawDebugPhase(RenderCommandBuffer& buffer, uint64_t submit_id,
-                              const DrawCallInfo& draw, uint32_t phase) {
+static void SetDrawDebugPhase(CommandBuffer& buffer, uint64_t submit_id, const DrawCallInfo& draw,
+                              uint32_t phase) {
 	EXIT_IF(draw.name == nullptr);
 
 	buffer.SetDebugInfo(static_cast<uint32_t>(draw.debug_op), submit_id, phase, draw.index_count,
@@ -957,8 +916,8 @@ static bool GetDrawTopology(const HW::UserConfig& ucfg, bool auto_draw,
 	return true;
 }
 
-static bool ResolvePrimitiveRestart(const RenderCommandBuffer& buffer,
-                                    vk::PrimitiveTopology topology, uint32_t index_type_and_size) {
+static bool ResolvePrimitiveRestart(const CommandBuffer& buffer, vk::PrimitiveTopology topology,
+                                    uint32_t index_type_and_size) {
 	const auto control = buffer.GetUserConfig().GetPrimitiveResetControl();
 	EXIT_NOT_IMPLEMENTED((control & ~0x3u) != 0);
 	if ((control & 0x1u) == 0) {
@@ -992,7 +951,7 @@ static bool ResolvePrimitiveRestart(const RenderCommandBuffer& buffer,
 	return true;
 }
 
-bool RenderExecutor::PrepareDrawRenderState(uint64_t submit_id, RenderCommandBuffer& buffer,
+bool RenderExecutor::PrepareDrawRenderState(uint64_t submit_id, CommandBuffer& buffer,
                                             const DrawCallInfo& draw,
                                             uint32_t            render_target_slice_offset,
                                             bool log_setup_phases, DrawRenderState& state) {
@@ -1032,7 +991,7 @@ bool RenderExecutor::PrepareDrawRenderState(uint64_t submit_id, RenderCommandBuf
 	return true;
 }
 
-static void RefreshShaders(RenderCommandBuffer& buffer, const DrawCallInfo& draw, bool log_phases,
+static void RefreshShaders(CommandBuffer& buffer, const DrawCallInfo& draw, bool log_phases,
                            DrawRenderState& state) {
 	EXIT_IF(draw.name == nullptr);
 	auto& ctx    = buffer.GetRegisters();
@@ -1042,37 +1001,33 @@ static void RefreshShaders(RenderCommandBuffer& buffer, const DrawCallInfo& draw
 	const auto& pixel_shader_info  = sh_ctx.GetPs();
 	const auto& shader_regs        = ctx.GetShaderRegisters();
 
-	state.vs_shader     = {};
-	state.ps_shader     = {};
-	state.ps_input_info = {};
+	state.vertex_program = {};
+	state.pixel_program  = {};
+	state.ps_input_info  = {};
 	std::array<Prospero::ColorComponentMapping, RENDER_COLOR_ATTACHMENTS_MAX>
 	    target_export_mapping {};
 	for (uint32_t i = 0; i < state.color_count; i++) {
 		target_export_mapping[state.color_info[i].target_slot] = state.color_info[i].export_mapping;
 	}
-	const auto lane_mask_mode = SelectGraphicsLaneMaskMode(64u);
-
 	if (log_phases) {
-		LogDrawPhase(draw.name, "ShaderCompileInfoVS");
-	}
-	if (!ShaderCompileInfoVS(vertex_shader_info, shader_regs, lane_mask_mode, state.vs_input_info,
-	                         state.vs_shader)) {
-		EXIT("ShaderCompileInfoVS failed for draw %s\n", draw.name);
+		LogDrawPhase(draw.name, "GetVertexProgram");
 	}
+	auto& pipeline_cache = buffer.GetContext().GetPipelineCache();
+	state.vertex_program =
+	    pipeline_cache.GetVertexProgram(vertex_shader_info, shader_regs, state.vs_input_info);
 
 	if (!state.ps_active) {
 		return;
 	}
 	if (log_phases) {
-		LogDrawPhase(draw.name, "ShaderCompileInfoPS");
-	}
-	if (!ShaderCompileInfoPS(pixel_shader_info, shader_regs, lane_mask_mode, state.vs_input_info,
-	                         target_export_mapping, state.ps_input_info, state.ps_shader)) {
-		EXIT("ShaderCompileInfoPS failed for draw %s\n", draw.name);
+		LogDrawPhase(draw.name, "GetPixelProgram");
 	}
+	state.pixel_program =
+	    pipeline_cache.GetPixelProgram(pixel_shader_info, shader_regs, state.vs_input_info,
+	                                   target_export_mapping, state.ps_input_info);
 }
 
-static PreparedVertexBuffers PrepareVertexBuffers(uint64_t submit_id, RenderCommandBuffer& buffer,
+static PreparedVertexBuffers PrepareVertexBuffers(uint64_t submit_id, CommandBuffer& buffer,
                                                   const DrawCallInfo&          draw,
                                                   const ShaderVertexInputInfo& vs_input_info) {
 	EXIT_IF(draw.name == nullptr);
@@ -1082,7 +1037,7 @@ static PreparedVertexBuffers PrepareVertexBuffers(uint64_t submit_id, RenderComm
 	return AcquireVertexBuffers(buffer, vs_input_info);
 }
 
-static PreparedIndexBuffer PrepareIndexBuffer(RenderCommandBuffer&         buffer,
+static PreparedIndexBuffer PrepareIndexBuffer(CommandBuffer&               buffer,
                                               const DrawIndexBufferSource& source) {
 	PreparedIndexBuffer prepared;
 	if (!source.enabled) {
@@ -1092,28 +1047,29 @@ static PreparedIndexBuffer PrepareIndexBuffer(RenderCommandBuffer&         buffe
 	prepared.size = source.size;
 	prepared.type = source.type;
 	if (source.host_data != nullptr) {
-		auto binding =
-		    buffer.GetContext().GetBufferCache().UploadTransient(source.host_data, source.size, 16);
-		prepared.owner  = std::move(binding.owner);
-		prepared.buffer = binding.buffer;
-		prepared.offset = binding.offset;
+		auto& stream = buffer.GetContext().GetBufferCache().GetUtilityBuffer(MemoryUsage::Stream);
+		prepared.offset = stream.Copy(source.host_data, source.size, 16);
+		prepared.buffer = stream.Handle();
 	} else {
-		auto binding =
-		    buffer.GetContext().GetBufferCache().ObtainBuffer(buffer, source.address, source.size);
-		prepared.owner  = std::move(binding.owner);
-		prepared.buffer = binding.buffer;
-		prepared.offset = binding.offset;
+		auto [buffer_ptr, offset] =
+		    buffer.GetContext().GetBufferCache().ObtainBuffer(source.address, source.size, false);
+		prepared.buffer = buffer_ptr->Handle();
+		prepared.offset = offset;
+	}
+	if (source.host_data != nullptr) {
+		SetVulkanObjectNameF(buffer.GetContext().GetGraphics().device, prepared.buffer,
+		                     "Kyty.IndexBuffer[guest=transient size=0x{:x} type={}]", source.size,
+		                     static_cast<uint32_t>(source.type));
+	} else {
+		SetVulkanObjectNameF(buffer.GetContext().GetGraphics().device, prepared.buffer,
+		                     "Kyty.IndexBuffer[guest=0x{:016x} size=0x{:x} type={}]",
+		                     source.address, source.size, static_cast<uint32_t>(source.type));
 	}
 	return prepared;
 }
 
-static void CommitVertexBuffers(RenderCommandBuffer& buffer, vk::CommandBuffer vk_buffer,
-                                PreparedVertexBuffers& prepared) {
-	for (uint32_t i = 0; i < prepared.owner_count; i++) {
-		if (prepared.owners[i] != nullptr) {
-			buffer.RetainResourceUntilFence(std::move(prepared.owners[i]));
-		}
-	}
+static void CommitVertexBuffers(vk::CommandBuffer            vk_buffer,
+                                const PreparedVertexBuffers& prepared) {
 	for (uint32_t i = 0; i < prepared.count; i++) {
 		EXIT_IF(prepared.buffers[i] == nullptr);
 	}
@@ -1123,19 +1079,15 @@ static void CommitVertexBuffers(RenderCommandBuffer& buffer, vk::CommandBuffer v
 	}
 }
 
-static void CommitIndexBuffer(RenderCommandBuffer& buffer, vk::CommandBuffer vk_buffer,
-                              PreparedIndexBuffer& prepared) {
+static void CommitIndexBuffer(vk::CommandBuffer vk_buffer, const PreparedIndexBuffer& prepared) {
 	if (prepared.size == 0) {
 		return;
 	}
-	if (prepared.owner != nullptr) {
-		buffer.RetainResourceUntilFence(std::move(prepared.owner));
-	}
 	EXIT_IF(prepared.buffer == nullptr);
 	vk_buffer.bindIndexBuffer(prepared.buffer, prepared.offset, prepared.type);
 }
 
-static void LogDrawStateIfNeeded(const RenderCommandBuffer& buffer, const DrawCallInfo& draw,
+static void LogDrawStateIfNeeded(const CommandBuffer& buffer, const DrawCallInfo& draw,
                                  const DrawRenderState& state, bool always_log,
                                  bool force_legacy_rect_log, uint32_t index_type_and_size,
                                  const void* index_addr) {
@@ -1203,7 +1155,7 @@ static void EmitDrawPrimitives(const HW::UserConfig& ucfg, vk::CommandBuffer vk_
 	}
 }
 
-void RenderExecutor::ExecutePreparedDraw(uint64_t submit_id, RenderCommandBuffer& buffer,
+void RenderExecutor::ExecutePreparedDraw(uint64_t submit_id, CommandBuffer& buffer,
                                          const DrawCallInfo& draw, DrawRenderState& state,
                                          vk::PrimitiveTopology topology, const DrawEmitInfo& emit,
                                          const DrawIndexBufferSource& index_source,
@@ -1213,8 +1165,8 @@ void RenderExecutor::ExecutePreparedDraw(uint64_t submit_id, RenderCommandBuffer
 	auto& ucfg = buffer.GetUserConfig();
 
 	LogDrawPhase(draw.name, "PrepareBindings");
-	auto bindings        = PrepareGraphicsBindings(buffer, state.vs_input_info.stage,
-	                                               state.ps_input_info.stage, state.ps_active);
+	auto bindings = PrepareGraphicsBindings(state.vs_input_info.stage, state.ps_input_info.stage,
+	                                        state.ps_active);
 	auto vertex_bindings = PrepareVertexBuffers(submit_id, buffer, draw, state.vs_input_info);
 	auto index_binding   = PrepareIndexBuffer(buffer, index_source);
 	state.rendering =
@@ -1224,9 +1176,9 @@ void RenderExecutor::ExecutePreparedDraw(uint64_t submit_id, RenderCommandBuffer
 		LogDrawPhase(draw.name, "CreatePipeline");
 	}
 	auto& pipeline = m_context.GetPipelineCache().CreateGraphicsPipeline(
-	    state.color_info, state.color_count, state.depth_info, state.vs_input_info, buffer,
-	    &state.ps_input_info, topology, primitive_restart_enable, state.ps_active, state.vs_shader,
-	    state.ps_shader);
+	    std::span {state.color_info, state.color_count}, state.depth_info, state.vs_input_info, buffer,
+	    state.ps_active ? &state.ps_input_info : nullptr, topology, primitive_restart_enable,
+	    state.vertex_program, state.pixel_program);
 
 	// Resource preparation above may synchronously finish and restart the scheduler. From this
 	// point onward, every operation targets the current command buffer and cannot touch guest
@@ -1238,21 +1190,23 @@ void RenderExecutor::ExecutePreparedDraw(uint64_t submit_id, RenderCommandBuffer
 	if (set_auto_debug) {
 		SetDrawDebugPhase(buffer, submit_id, draw, 0x200u);
 	}
-	CommitVertexBuffers(buffer, vk_buffer, vertex_bindings);
-	CommitBindings(buffer, vk::PipelineBindPoint::eGraphics, pipeline.pipeline_layout,
-	               bindings.vertex);
+	CommitVertexBuffers(vk_buffer, vertex_bindings);
 	if (bindings.pixel.has_value()) {
 		if (set_auto_debug) {
 			SetDrawDebugPhase(buffer, submit_id, draw, 0x300u);
 		}
-		CommitBindings(buffer, vk::PipelineBindPoint::eGraphics, pipeline.pipeline_layout,
-		               *bindings.pixel);
 	}
-	CommitIndexBuffer(buffer, vk_buffer, index_binding);
+	std::array<PreparedBindings*, 2> descriptor_stages {&bindings.vertex, nullptr};
+	const size_t                     descriptor_stage_count = bindings.pixel.has_value() ? 2u : 1u;
+	if (bindings.pixel) {
+		descriptor_stages[1] = &*bindings.pixel;
+	}
+	CommitBindings(buffer, vk::PipelineBindPoint::eGraphics, pipeline,
+	               std::span {descriptor_stages.data(), descriptor_stage_count});
+	CommitIndexBuffer(vk_buffer, index_binding);
 
-	const auto dynamic_params =
-	    BuildGraphicsDynamicParams(buffer, state.color_info, state.color_count, state.depth_info);
-	SetDynamicParams(buffer, vk_buffer, dynamic_params);
+	SetGraphicsDynamicParams(buffer, vk_buffer, state.color_info, state.color_count,
+	                         state.depth_info);
 
 	LogDrawPhase(draw.name, "BeginRendering");
 	if (set_auto_debug) {
@@ -1285,7 +1239,7 @@ void RenderExecutor::ExecutePreparedDraw(uint64_t submit_id, RenderCommandBuffer
 	}
 }
 
-void RenderExecutor::DrawIndex(uint64_t submit_id, RenderCommandBuffer& buffer,
+void RenderExecutor::DrawIndex(uint64_t submit_id, CommandBuffer& buffer,
                                uint32_t index_type_and_size, uint32_t index_count,
                                const void* index_addr, uint32_t flags, uint32_t type,
                                uint32_t instance_count, uint32_t render_target_slice_offset,
@@ -1293,6 +1247,7 @@ void RenderExecutor::DrawIndex(uint64_t submit_id, RenderCommandBuffer& buffer,
 	KYTY_PROFILER_FUNCTION();
 
 	EXIT_IF(buffer.IsInvalid());
+	m_context.GetCommandScheduler().PopPendingOperations();
 	auto& ucfg   = buffer.GetUserConfig();
 	auto& sh_ctx = buffer.GetShaders();
 
@@ -1417,13 +1372,14 @@ void RenderExecutor::DrawIndex(uint64_t submit_id, RenderCommandBuffer& buffer,
 }
 
 // NOLINTNEXTLINE(readability-function-cognitive-complexity)
-void RenderExecutor::DrawAuto(uint64_t submit_id, RenderCommandBuffer& buffer, uint32_t index_count,
+void RenderExecutor::DrawAuto(uint64_t submit_id, CommandBuffer& buffer, uint32_t index_count,
                               uint32_t flags, uint32_t render_target_slice_offset,
                               uint32_t instance_count, uint32_t first_vertex,
                               uint32_t first_instance) {
 	KYTY_PROFILER_FUNCTION();
 
 	EXIT_IF(buffer.IsInvalid());
+	m_context.GetCommandScheduler().PopPendingOperations();
 	auto& ucfg   = buffer.GetUserConfig();
 	auto& sh_ctx = buffer.GetShaders();
 
@@ -1493,7 +1449,7 @@ void RenderExecutor::DrawAuto(uint64_t submit_id, RenderCommandBuffer& buffer, u
 		if (graphics_debug_dump_enabled()) {
 			LOGF("DrawIndexAuto: skipping rect-list draw with no VS param exports and PS inputs: "
 			     "ps_inputs=%u ps=0x%016" PRIx64 " es=0x%016" PRIx64 " gs=0x%016" PRIx64 "\n",
-			     state.ps_input_info.input_num, sh_ctx.GetPs().ps_regs.chksum,
+			     state.ps_input_info.input_num, sh_ctx.GetPs().ps_regs.data_addr,
 			     sh_ctx.GetVs().es_regs.data_addr, sh_ctx.GetVs().gs_regs.data_addr);
 		}
 		ResetBindings();
@@ -1515,7 +1471,7 @@ void RenderExecutor::DrawAuto(uint64_t submit_id, RenderCommandBuffer& buffer, u
 	ResetBindings();
 }
 
-bool RenderExecutor::ResolveColorTargets(uint64_t submit_id, RenderCommandBuffer& buffer,
+bool RenderExecutor::ResolveColorTargets(uint64_t submit_id, CommandBuffer& buffer,
                                          uint32_t render_target_slice_offset) {
 	const auto& hw = buffer.GetRegisters();
 	if (hw.GetColorControl().mode != 3) {

src/graphics/host_gpu/renderer/renderDraw.h

@@ -1,43 +1,12 @@
 #ifndef EMULATOR_SRC_GRAPHICS_HOST_GPU_RENDERER_RENDERDRAW_H_
 #define EMULATOR_SRC_GRAPHICS_HOST_GPU_RENDERER_RENDERDRAW_H_
 
-#include "graphics/host_gpu/renderer/renderTarget.h"
-
 #include <cstdint>
-#include <type_traits>
 
 namespace Libs::Graphics {
 
-struct RenderColorInfo;
 struct ShaderVertexInputInfo;
 
-#pragma pack(push, 1)
-
-struct PipelineDynamicParameters {
-	bool stencil_test_enable = false;
-	bool depth_bias_enable   = false;
-
-	float viewport_scale[3]  = {};
-	float viewport_offset[3] = {};
-	int   scissor_ltrb[4]    = {0};
-
-	float    line_width                                       = 1.0f;
-	float    depth_bias_constant_factor                       = 0.0f;
-	float    depth_bias_clamp                                 = 0.0f;
-	float    depth_bias_slope_factor                          = 0.0f;
-	uint32_t color_write_count                                = 1;
-	bool     color_write_enable[RENDER_COLOR_ATTACHMENTS_MAX] = {true, true};
-
-	PipelineStencilDynamicState stencil_front;
-	PipelineStencilDynamicState stencil_back;
-};
-
-#pragma pack(pop)
-
-static_assert(std::is_trivially_copyable_v<PipelineDynamicParameters>);
-static_assert(std::is_standard_layout_v<PipelineDynamicParameters>);
-static_assert(alignof(PipelineDynamicParameters) == 1);
-
 [[nodiscard]] int32_t ResolveVertexOffset(uint32_t                     index_offset,
                                           const ShaderVertexInputInfo& vs_input_info);
 

src/graphics/host_gpu/renderer/sync.cpp

@@ -20,9 +20,7 @@
 
 namespace Libs::Graphics::Sync {
 
-constexpr int      GRAPHICS_EVENT_QUEUED_GRAPHICS_INTERRUPT = 0x00;
-constexpr int      GRAPHICS_EVENT_EOP                       = 0x40;
-constexpr uint64_t GRAPHICS_REFERENCE_CLOCK_FREQUENCY       = 100000000;
+constexpr uint64_t GRAPHICS_REFERENCE_CLOCK_FREQUENCY = 100000000;
 
 bool ScaleReferenceClock(uint64_t host_ticks, uint64_t host_frequency, uint64_t& value) {
 	if (host_frequency == 0) {
@@ -66,8 +64,9 @@ struct EndOfPipeSignal {
 	std::array<uint32_t, 4> debug_args      = {};
 	uint64_t                debug_data      = 0;
 	std::optional<uint64_t> destination;
-	EndOfPipeCompletion     completion      = EndOfPipeCompletion::None;
-	uint64_t                completion_data = 0;
+	EndOfPipeCompletion     completion         = EndOfPipeCompletion::None;
+	uint64_t                completion_data    = 0;
+	int                     interrupt_event_id = 0;
 };
 
 enum class EndOfPipeWriteSize : uint32_t { Dword = 4, Qword = 8 };
@@ -96,8 +95,10 @@ static void RecordEndOfPipeSignal(const EndOfPipeSignal& signal) {
 		case EndOfPipeCompletion::None: return;
 		case EndOfPipeCompletion::Interrupt: {
 			const auto context_id = static_cast<uint32_t>(signal.completion_data);
-			scheduler.DeferPriorityOperation(
-			    [&renderer, context_id] { renderer.TriggerEopEvent(context_id); });
+			const auto event_id   = signal.interrupt_event_id;
+			scheduler.DeferPriorityOperation([&renderer, event_id, context_id] {
+				renderer.TriggerInterrupt(event_id, context_id);
+			});
 			return;
 		}
 		case EndOfPipeCompletion::Flip: {
@@ -108,9 +109,10 @@ static void RecordEndOfPipeSignal(const EndOfPipeSignal& signal) {
 		}
 		case EndOfPipeCompletion::FlipAndInterrupt: {
 			const auto request_id = signal.completion_data;
-			scheduler.DeferPriorityOperation([&renderer, request_id] {
+			const auto event_id   = signal.interrupt_event_id;
+			scheduler.DeferPriorityOperation([&renderer, event_id, request_id] {
 				renderer.GetVideoOut().CompleteFlip(request_id);
-				renderer.TriggerEopEvent(0);
+				renderer.TriggerInterrupt(event_id, 0);
 			});
 			return;
 		}
@@ -135,7 +137,8 @@ static bool TriggersInterrupt(EndOfPipeWriteAction action) {
 
 static void RecordEndOfPipeWrite(uint64_t submit_id, CommandBuffer& buffer, uint64_t destination,
                                  uint64_t value, EndOfPipeWriteSize size,
-                                 EndOfPipeWriteAction action, uint32_t context_id = 0) {
+                                 EndOfPipeWriteAction action, int interrupt_event_id = 0,
+                                 uint32_t context_id = 0) {
 	const auto width      = static_cast<uint32_t>(size);
 	const auto value_low  = static_cast<uint32_t>(value);
 	const auto value_high = static_cast<uint32_t>(value >> 32u);
@@ -150,18 +153,12 @@ static void RecordEndOfPipeWrite(uint64_t submit_id, CommandBuffer& buffer, uint
 	    .debug_data      = destination,
 	    .destination     = destination,
 	    .completion      = interrupt ? EndOfPipeCompletion::Interrupt : EndOfPipeCompletion::None,
-	    .completion_data = context_id != 0 ? context_id : value,
+	    .completion_data = interrupt ? context_id : value,
+	    .interrupt_event_id = interrupt_event_id,
 	};
 	RecordEndOfPipeSignal(signal);
 }
 
-void TriggerAgcUserInterrupt() {
-	auto tsc    = LibKernel::KernelReadTsc();
-	auto result = LibKernel::EventQueue::KernelTriggerUserEventForAll(AGC_USER_INTERRUPT_EVENT,
-	                                                                  reinterpret_cast<void*>(tsc));
-	EXIT_NOT_IMPLEMENTED(result != OK && result != LibKernel::KERNEL_ERROR_ENOENT);
-}
-
 void WriteAtEndOfPipe32(uint64_t submit_id, CommandBuffer& buffer, uint32_t* dst_gpu_addr,
                         uint32_t value) {
 	RecordEndOfPipeWrite(submit_id, buffer, reinterpret_cast<uint64_t>(dst_gpu_addr), value,
@@ -220,31 +217,35 @@ void WriteAtEndOfPipeWithWriteBack32(uint64_t submit_id, CommandBuffer& buffer,
 }
 
 void WriteAtEndOfPipeWithInterruptWriteBack64(uint64_t submit_id, CommandBuffer& buffer,
-                                              uint64_t* dst_gpu_addr, uint64_t value,
+                                              uint64_t* dst_gpu_addr, uint64_t value, int event_id,
                                               uint32_t context_id) {
 	RecordEndOfPipeWrite(submit_id, buffer, reinterpret_cast<uint64_t>(dst_gpu_addr), value,
 	                     EndOfPipeWriteSize::Qword, EndOfPipeWriteAction::InterruptWriteBack,
-	                     context_id);
+	                     event_id, context_id);
 }
 
 void WriteAtEndOfPipeWithInterruptWriteBack32(uint64_t submit_id, CommandBuffer& buffer,
-                                              uint32_t* dst_gpu_addr, uint32_t value,
+                                              uint32_t* dst_gpu_addr, uint32_t value, int event_id,
                                               uint32_t context_id) {
 	RecordEndOfPipeWrite(submit_id, buffer, reinterpret_cast<uint64_t>(dst_gpu_addr), value,
 	                     EndOfPipeWriteSize::Dword, EndOfPipeWriteAction::InterruptWriteBack,
-	                     context_id);
+	                     event_id, context_id);
 }
 
 void WriteAtEndOfPipeWithInterrupt64(uint64_t submit_id, CommandBuffer& buffer,
-                                     uint64_t* dst_gpu_addr, uint64_t value, uint32_t context_id) {
+                                     uint64_t* dst_gpu_addr, uint64_t value, int event_id,
+                                     uint32_t context_id) {
 	RecordEndOfPipeWrite(submit_id, buffer, reinterpret_cast<uint64_t>(dst_gpu_addr), value,
-	                     EndOfPipeWriteSize::Qword, EndOfPipeWriteAction::Interrupt, context_id);
+	                     EndOfPipeWriteSize::Qword, EndOfPipeWriteAction::Interrupt, event_id,
+	                     context_id);
 }
 
 void WriteAtEndOfPipeWithInterrupt32(uint64_t submit_id, CommandBuffer& buffer,
-                                     uint32_t* dst_gpu_addr, uint32_t value, uint32_t context_id) {
+                                     uint32_t* dst_gpu_addr, uint32_t value, int event_id,
+                                     uint32_t context_id) {
 	RecordEndOfPipeWrite(submit_id, buffer, reinterpret_cast<uint64_t>(dst_gpu_addr), value,
-	                     EndOfPipeWriteSize::Dword, EndOfPipeWriteAction::Interrupt, context_id);
+	                     EndOfPipeWriteSize::Dword, EndOfPipeWriteAction::Interrupt, event_id,
+	                     context_id);
 }
 
 uint64_t PrepareVideoOutFlip(CommandBuffer& buffer, int handle, int index, int flip_mode,
@@ -270,18 +271,20 @@ uint64_t PrepareVideoOutFlip(CommandBuffer& buffer, int handle, int index, int f
 void WriteAtEndOfPipeWithInterruptWriteBackFlip32(uint64_t submit_id, CommandBuffer& buffer,
                                                   uint32_t* dst_gpu_addr, uint32_t value,
                                                   int handle, int index, int flip_mode,
-                                                  int64_t flip_arg, uint64_t request_id) {
+                                                  int64_t flip_arg, uint64_t request_id,
+                                                  int event_id) {
 	const auto destination = reinterpret_cast<uint64_t>(dst_gpu_addr);
 	RecordEndOfPipeSignal({
-	    .buffer          = &buffer,
-	    .submit_id       = submit_id,
-	    .debug_operation = CommandBufferDebugOp::EopWriteBackFlip,
-	    .debug_args      = {static_cast<uint32_t>(handle), static_cast<uint32_t>(index),
-	                        static_cast<uint32_t>(flip_mode), value},
-	    .debug_data      = static_cast<uint64_t>(flip_arg),
-	    .destination     = destination,
-	    .completion      = EndOfPipeCompletion::FlipAndInterrupt,
-	    .completion_data = request_id,
+	    .buffer             = &buffer,
+	    .submit_id          = submit_id,
+	    .debug_operation    = CommandBufferDebugOp::EopWriteBackFlip,
+	    .debug_args         = {static_cast<uint32_t>(handle), static_cast<uint32_t>(index),
+	                           static_cast<uint32_t>(flip_mode), value},
+	    .debug_data         = static_cast<uint64_t>(flip_arg),
+	    .destination        = destination,
+	    .completion         = EndOfPipeCompletion::FlipAndInterrupt,
+	    .completion_data    = request_id,
+	    .interrupt_event_id = event_id,
 	});
 }
 
@@ -316,36 +319,24 @@ void WriteAtEndOfPipeOnlyFlip(uint64_t submit_id, CommandBuffer& buffer, int han
 	});
 }
 
-void TriggerEopEventAtEndOfPipe(CommandBuffer& buffer, uint32_t context_id) {
+void TriggerEopEventAtEndOfPipe(CommandBuffer& buffer, int event_id, uint32_t context_id) {
 	ValidateEndOfPipeSignal({.buffer = &buffer});
 	auto& renderer  = buffer.GetContext();
 	auto& scheduler = renderer.GetCommandScheduler();
 	EXIT_IF(!scheduler.Active() || &buffer != &scheduler.Current());
 	scheduler.DeferPriorityOperation(
-	    [&renderer, context_id] { renderer.TriggerEopEvent(context_id); });
+	    [&renderer, event_id, context_id] { renderer.TriggerInterrupt(event_id, context_id); });
 }
 
-static void EopEventResetFunc(LibKernel::EventQueue::KernelEqueueEvent* event) {
+static void InterruptEventResetFunc(LibKernel::EventQueue::KernelEqueueEvent* event) {
 	EXIT_IF(event == nullptr);
 	event->triggered    = false;
 	event->event.fflags = 0;
 	event->event.data   = 0;
 }
 
-static void EopEventDeleteFunc(LibKernel::EventQueue::KernelEqueue       eq,
-                               LibKernel::EventQueue::KernelEqueueEvent* event) {
-	EXIT_IF(event == nullptr);
-	EXIT_NOT_IMPLEMENTED(event->event.filter != LibKernel::EventQueue::KERNEL_EVFILT_GRAPHICS);
-	if (event->event.ident == GRAPHICS_EVENT_QUEUED_GRAPHICS_INTERRUPT ||
-	    event->event.ident == GRAPHICS_EVENT_EOP) {
-		auto* renderer = static_cast<RenderContext*>(event->filter.data);
-		EXIT_IF(renderer == nullptr);
-		renderer->DeleteEopEq(eq, static_cast<int>(event->event.ident));
-	}
-}
-
-static void EopEventTriggerFunc(LibKernel::EventQueue::KernelEqueueEvent* event,
-                                void*                                     trigger_data) {
+static void InterruptEventTriggerFunc(LibKernel::EventQueue::KernelEqueueEvent* event,
+                                      void*                                     trigger_data) {
 	EXIT_IF(event == nullptr);
 
 	auto triggered_event = event->event;
@@ -362,35 +353,35 @@ static void EopEventTriggerFunc(LibKernel::EventQueue::KernelEqueueEvent* event,
 int AddEqEvent(RenderContext& renderer, LibKernel::EventQueue::KernelEqueue eq, int id,
                void* udata) {
 	LibKernel::EventQueue::KernelEqueueEvent event;
-	event.triggered                = false;
-	event.event.ident              = static_cast<uintptr_t>(id);
-	event.event.filter             = LibKernel::EventQueue::KERNEL_EVFILT_GRAPHICS;
-	event.event.udata              = udata;
-	event.event.fflags             = 0;
-	event.event.data               = id;
-	event.filter.delete_event_func = EopEventDeleteFunc;
-	event.filter.reset_func        = EopEventResetFunc;
-	event.filter.trigger_func      = EopEventTriggerFunc;
-	event.filter.data              = &renderer;
+	event.triggered           = false;
+	event.event.ident         = static_cast<uintptr_t>(id);
+	event.event.filter        = LibKernel::EventQueue::KERNEL_EVFILT_GRAPHICS;
+	event.event.udata         = udata;
+	event.event.fflags        = 0;
+	event.event.data          = id;
+	event.filter.reset_func   = InterruptEventResetFunc;
+	event.filter.trigger_func = InterruptEventTriggerFunc;
 
 	int result = LibKernel::EventQueue::KernelAddEvent(eq, event);
 
-	if (result == 0 &&
-	    (id == GRAPHICS_EVENT_QUEUED_GRAPHICS_INTERRUPT || id == GRAPHICS_EVENT_EOP)) {
-		renderer.AddEopEq(eq, id);
+	if (result == 0) {
+		renderer.AddInterruptEq(eq, id);
 	}
 
 	return result;
 }
 
-int DeleteEqEvent(LibKernel::EventQueue::KernelEqueue eq, int id) {
+int DeleteEqEvent(RenderContext& renderer, LibKernel::EventQueue::KernelEqueue eq, int id) {
 	int result = LibKernel::EventQueue::KernelDeleteEvent(
 	    eq, static_cast<uintptr_t>(id), LibKernel::EventQueue::KERNEL_EVFILT_GRAPHICS);
+	if (result == OK || result == LibKernel::KERNEL_ERROR_ENOENT) {
+		renderer.DeleteInterruptEq(eq, id);
+	}
 
 	return result;
 }
 
-void ReadGds(Buffer& gds, uint32_t* dst, uint32_t dw_offset, uint32_t dw_size) {
+void ReadGds(const Buffer& gds, uint32_t* dst, uint32_t dw_offset, uint32_t dw_size) {
 	const auto offset = uint64_t {dw_offset} * sizeof(uint32_t);
 	const auto size   = uint64_t {dw_size} * sizeof(uint32_t);
 	EXIT_IF(dst == nullptr || offset > gds.Size() || size > gds.Size() - offset ||

src/graphics/host_gpu/renderer/sync.h

@@ -17,8 +17,7 @@ namespace Sync {
                                            uint64_t& value);
 [[nodiscard]] uint64_t ReadReferenceClock();
 
-void TriggerAgcUserInterrupt();
-void TriggerEopEventAtEndOfPipe(CommandBuffer& buffer, uint32_t context_id);
+void TriggerEopEventAtEndOfPipe(CommandBuffer& buffer, int event_id, uint32_t context_id);
 
 void WriteAtEndOfPipe32(uint64_t submit_id, CommandBuffer& buffer, uint32_t* dst_gpu_addr,
                         uint32_t value);
@@ -35,16 +34,16 @@ void WriteAtEndOfPipeWithWriteBack32(uint64_t submit_id, CommandBuffer& buffer,
 void WriteAtEndOfPipeWithWriteBack64(uint64_t submit_id, CommandBuffer& buffer,
                                      uint64_t* dst_gpu_addr, uint64_t value);
 void WriteAtEndOfPipeWithInterrupt32(uint64_t submit_id, CommandBuffer& buffer,
-                                     uint32_t* dst_gpu_addr, uint32_t value,
+                                     uint32_t* dst_gpu_addr, uint32_t value, int event_id,
                                      uint32_t context_id = 0);
 void WriteAtEndOfPipeWithInterrupt64(uint64_t submit_id, CommandBuffer& buffer,
-                                     uint64_t* dst_gpu_addr, uint64_t value,
+                                     uint64_t* dst_gpu_addr, uint64_t value, int event_id,
                                      uint32_t context_id = 0);
 void WriteAtEndOfPipeWithInterruptWriteBack32(uint64_t submit_id, CommandBuffer& buffer,
-                                              uint32_t* dst_gpu_addr, uint32_t value,
+                                              uint32_t* dst_gpu_addr, uint32_t value, int event_id,
                                               uint32_t context_id = 0);
 void WriteAtEndOfPipeWithInterruptWriteBack64(uint64_t submit_id, CommandBuffer& buffer,
-                                              uint64_t* dst_gpu_addr, uint64_t value,
+                                              uint64_t* dst_gpu_addr, uint64_t value, int event_id,
                                               uint32_t context_id = 0);
 
 [[nodiscard]] uint64_t PrepareVideoOutFlip(CommandBuffer& buffer, int handle, int index,
@@ -57,12 +56,13 @@ void WriteAtEndOfPipeWithFlip32(uint64_t submit_id, CommandBuffer& buffer, uint3
 void WriteAtEndOfPipeWithInterruptWriteBackFlip32(uint64_t submit_id, CommandBuffer& buffer,
                                                   uint32_t* dst_gpu_addr, uint32_t value,
                                                   int handle, int index, int flip_mode,
-                                                  int64_t flip_arg, uint64_t request_id);
+                                                  int64_t flip_arg, uint64_t request_id,
+                                                  int event_id);
 
 int  AddEqEvent(RenderContext& renderer, LibKernel::EventQueue::KernelEqueue eq, int id,
                 void* udata);
-int  DeleteEqEvent(LibKernel::EventQueue::KernelEqueue eq, int id);
-void ReadGds(Buffer& gds, uint32_t* dst, uint32_t dw_offset, uint32_t dw_size);
+int  DeleteEqEvent(RenderContext& renderer, LibKernel::EventQueue::KernelEqueue eq, int id);
+void ReadGds(const Buffer& gds, uint32_t* dst, uint32_t dw_offset, uint32_t dw_size);
 
 } // namespace Sync
 } // namespace Libs::Graphics

src/graphics/host_gpu/shaders/fault_buffer_process.comp

@@ -0,0 +1,35 @@
+// SPDX-FileCopyrightText: Copyright 2024 shadPS4 Emulator Project
+// SPDX-License-Identifier: GPL-2.0-or-later
+
+#version 450
+#extension GL_ARB_gpu_shader_int64 : enable
+
+layout(local_size_x = 64, local_size_y = 1, local_size_z = 1) in;
+
+layout(std430, binding = 0) buffer input_buf {
+    uint fault_buffer[];
+};
+
+layout(std430, binding = 1) buffer output_buf {
+    uint64_t download_buffer[];
+};
+layout(std430, binding = 1) buffer output_buf32 {
+    uint download_buffer32[];
+};
+
+void main() {
+    const uint id = gl_GlobalInvocationID.x;
+    uint word = fault_buffer[id];
+    fault_buffer[id] = 0u;
+    const uint base_bit = id * 32u;
+    while (word != 0u) {
+        const uint store_index = atomicAdd(download_buffer32[0], 1u) + 1u;
+        if (store_index >= MAX_PAGE_FAULTS) {
+            return;
+        }
+        const uint bit = findLSB(word);
+        word &= word - 1;
+        const uint page = base_bit + bit;
+        download_buffer[store_index] = uint64_t(page) << CACHING_PAGEBITS;
+    }
+}

src/graphics/host_gpu/vma.cpp

@@ -72,7 +72,10 @@ bool GraphicContext::CreateAllocator() {
 	info.device           = device;
 	info.pVulkanFunctions = &functions;
 	info.vulkanApiVersion = VULKAN_TARGET_API_VERSION;
-	info.flags = memory_budget_ext_enabled ? VMA_ALLOCATOR_CREATE_EXT_MEMORY_BUDGET_BIT : 0;
+	info.flags = VMA_ALLOCATOR_CREATE_BUFFER_DEVICE_ADDRESS_BIT;
+	if (memory_budget_ext_enabled) {
+		info.flags |= VMA_ALLOCATOR_CREATE_EXT_MEMORY_BUDGET_BIT;
+	}
 
 	const auto result = static_cast<vk::Result>(vmaCreateAllocator(&info, &allocator));
 	if (result != vk::Result::eSuccess) {
@@ -199,20 +202,6 @@ void GraphicContext::CreateBuffer(uint64_t size, VulkanBuffer& buffer) {
 	VulkanTrackAllocation(buffer.memory);
 }
 
-void GraphicContext::DeleteBuffer(VulkanBuffer& buffer) {
-	KYTY_PROFILER_FUNCTION();
-	EXIT_IF(allocator == nullptr || buffer.buffer == nullptr ||
-	        buffer.memory.allocation == nullptr);
-
-	VulkanUntrackAllocation(buffer.memory);
-	vmaDestroyBuffer(allocator, buffer.buffer, buffer.memory.allocation);
-	buffer.buffer                 = nullptr;
-	buffer.memory.memory          = nullptr;
-	buffer.memory.allocation      = nullptr;
-	buffer.memory.allocation_info = {};
-	buffer.memory.offset          = 0;
-}
-
 bool GraphicContext::CreateImage(const vk::ImageCreateInfo& image_info, VulkanImage& image) {
 	KYTY_PROFILER_FUNCTION();
 	EXIT_IF(allocator == nullptr || image.image != nullptr || image.memory.allocation != nullptr);

src/graphics/host_gpu/vulkanCommon.h

@@ -10,10 +10,13 @@
 #define VMA_STATIC_VULKAN_FUNCTIONS  0
 #define VMA_DYNAMIC_VULKAN_FUNCTIONS 1
 
+#include "common/emulatorConfig.h"
 #include "graphics/guest_gpu/gpu_defs.h"
 
 #include <cstdint>
+#include <fmt/format.h>
 #include <string>
+#include <utility>
 #include <vector>
 #include <vulkan/vulkan.hpp>
 #include <vulkan/vulkan_hash.hpp>
@@ -29,6 +32,24 @@ std::string VulkanToString(vk::QueueFlags value);
 vk::Format  VulkanFormat(Prospero::BufferFormat guest_format);
 void        RequireVulkanSuccess(vk::Result result, const char* operation);
 
+template <typename Handle, typename... Args>
+void SetVulkanObjectNameF(vk::Device device, Handle handle, fmt::format_string<Args...> format,
+                          Args&&... args) {
+	if (!Config::GraphicsDebugDumpEnabled() || device == nullptr || handle == nullptr ||
+	    VULKAN_HPP_DEFAULT_DISPATCHER.vkSetDebugUtilsObjectNameEXT == nullptr) {
+		return;
+	}
+
+	const auto                      name = fmt::format(format, std::forward<Args>(args)...);
+	vk::DebugUtilsObjectNameInfoEXT info {};
+	info.sType        = vk::StructureType::eDebugUtilsObjectNameInfoEXT;
+	info.objectType   = Handle::objectType;
+	info.objectHandle = static_cast<uint64_t>(
+	    reinterpret_cast<uintptr_t>(static_cast<typename Handle::CType>(handle)));
+	info.pObjectName = name.c_str();
+	(void)device.setDebugUtilsObjectNameEXT(&info);
+}
+
 template <typename T, typename Enumerator>
 [[nodiscard]] std::vector<T> EnumerateVulkan(const char* operation, Enumerator&& enumerate) {
 	for (;;) {

src/graphics/host_gpu/vulkanInstance.h

@@ -1,93 +0,0 @@
-#ifndef EMULATOR_SRC_GRAPHICS_HOST_GPU_VULKANINSTANCE_H_
-#define EMULATOR_SRC_GRAPHICS_HOST_GPU_VULKANINSTANCE_H_
-
-#include "common/abi.h"
-#include "common/common.h"
-#include "common/threads.h"
-#include "graphics/host_gpu/vulkanCommon.h"
-
-#include <map>
-#include <mutex>
-#include <tuple>
-#include <vk_mem_alloc.h>
-
-namespace Libs::Graphics {
-
-inline constexpr uint32_t VULKAN_TARGET_API_VERSION = VK_API_VERSION_1_3;
-
-struct VulkanInstance {
-	vk::Instance                       instance                          = nullptr;
-	vk::DebugUtilsMessengerEXT         debug_messenger                   = nullptr;
-	vk::PhysicalDevice                 physical_device                   = nullptr;
-	vk::PhysicalDeviceProperties       physical_device_properties        = {};
-	vk::PhysicalDeviceMemoryProperties physical_device_memory_properties = {};
-	vk::Device                         device                            = nullptr;
-	VmaAllocator                       allocator                         = nullptr;
-	bool                               memory_budget_ext_enabled         = false;
-	bool                               rt_extensions_enabled             = false;
-	bool                               subgroup_size_control_enabled     = false;
-	bool                               sample_rate_shading_enabled       = false;
-	uint32_t                           subgroup_size                     = 0;
-	uint32_t                           min_subgroup_size                 = 0;
-	uint32_t                           max_subgroup_size                 = 0;
-	vk::ShaderStageFlags               required_subgroup_size_stages     = {};
-	Common::Mutex                      queue_mutex;
-	uint32_t                           queue_family = static_cast<uint32_t>(-1);
-	vk::Queue                          queue        = nullptr;
-
-	[[nodiscard]] const vk::PhysicalDeviceProperties& GetPhysicalDeviceProperties() const {
-		return physical_device_properties;
-	}
-
-	[[nodiscard]] const vk::PhysicalDeviceMemoryProperties&
-	GetPhysicalDeviceMemoryProperties() const {
-		return physical_device_memory_properties;
-	}
-
-	[[nodiscard]] vk::FormatProperties GetFormatProperties(vk::Format format) const {
-		std::scoped_lock lock(m_format_properties_mutex);
-		auto [it, inserted] = m_format_properties.try_emplace(format);
-		if (inserted) {
-			physical_device.getFormatProperties(format, &it->second);
-		}
-		return it->second;
-	}
-
-	[[nodiscard]] vk::Result GetImageFormatProperties(vk::Format format, vk::ImageType type,
-	                                                  vk::ImageTiling            tiling,
-	                                                  vk::ImageUsageFlags        usage,
-	                                                  vk::ImageCreateFlags       flags,
-	                                                  vk::ImageFormatProperties* properties) const {
-		using Key = std::tuple<vk::Format, vk::ImageType, vk::ImageTiling, vk::ImageUsageFlags,
-		                       vk::ImageCreateFlags>;
-		std::scoped_lock lock(m_image_format_properties_mutex);
-		auto [it, inserted] =
-		    m_image_format_properties.try_emplace(Key {format, type, tiling, usage, flags});
-		if (inserted) {
-			it->second.first = physical_device.getImageFormatProperties(format, type, tiling, usage,
-			                                                            flags, &it->second.second);
-		}
-		if (properties != nullptr) {
-			*properties = it->second.second;
-		}
-		return it->second.first;
-	}
-
-	[[nodiscard]] vk::DeviceSize StorageMinAlignment() const {
-		const auto alignment = physical_device_properties.limits.minStorageBufferOffsetAlignment;
-		return alignment != 0 ? alignment : 1;
-	}
-
-private:
-	mutable std::mutex                                 m_format_properties_mutex;
-	mutable std::map<vk::Format, vk::FormatProperties> m_format_properties;
-	mutable std::mutex                                 m_image_format_properties_mutex;
-	mutable std::map<std::tuple<vk::Format, vk::ImageType, vk::ImageTiling, vk::ImageUsageFlags,
-	                            vk::ImageCreateFlags>,
-	                 std::pair<vk::Result, vk::ImageFormatProperties>>
-	    m_image_format_properties;
-};
-
-} // namespace Libs::Graphics
-
-#endif // EMULATOR_SRC_GRAPHICS_HOST_GPU_VULKANINSTANCE_H_

src/graphics/presentation/window/hostInput.cpp

@@ -1,18 +1,24 @@
 #include "graphics/presentation/window/hostInput.h"
 
+#include "SDL_error.h"
+#include "SDL_events.h"
 #include "SDL_keyboard.h"
 #include "SDL_keycode.h"
 #include "SDL_mouse.h"
+#include "SDL_timer.h"
+#include "SDL_video.h"
 #include "common/assert.h"
 #include "common/emulatorConfig.h"
+#include "common/logging/log.h"
 #include "libs/controller.h"
 
 #include <algorithm>
 #include <array>
+#include <cmath>
 #include <cstddef>
+#include <cstdlib>
 #include <string>
 #include <string_view>
-#include <utility>
 
 namespace Libs::Graphics {
 
@@ -23,6 +29,7 @@ struct ControlInfo {
 	uint32_t         button   = 0;
 	Controller::Axis axis     = Controller::Axis::AxisMax;
 	bool             positive = false;
+	float            touch_x  = 0.0f;
 };
 
 static constexpr std::array CONTROL_INFO = {
@@ -41,7 +48,10 @@ static constexpr std::array CONTROL_INFO = {
     ControlInfo {"Circle", Controller::PAD_BUTTON_CIRCLE},
     ControlInfo {"Cross", Controller::PAD_BUTTON_CROSS},
     ControlInfo {"Square", Controller::PAD_BUTTON_SQUARE},
-    ControlInfo {"TouchPad", Controller::PAD_BUTTON_TOUCH_PAD},
+    ControlInfo {"TouchPad", Controller::PAD_BUTTON_TOUCH_PAD, Controller::Axis::AxisMax, false,
+                 0.25f},
+    ControlInfo {"TouchPadRight", Controller::PAD_BUTTON_TOUCH_PAD, Controller::Axis::AxisMax,
+                 false, 0.75f},
     ControlInfo {"LeftStickLeft", 0, Controller::Axis::LeftX, false},
     ControlInfo {"LeftStickRight", 0, Controller::Axis::LeftX, true},
     ControlInfo {"LeftStickUp", 0, Controller::Axis::LeftY, false},
@@ -60,17 +70,23 @@ struct Binding {
 	std::size_t control      = INVALID_CONTROL;
 };
 
+constexpr int              MOUSE_POLL_INTERVAL_MS = 33;
+constexpr std::string_view MOUSE_SENSITIVITY      = "MouseSensitivity=";
+
+struct MouseJoystickState {
+	bool     enabled   = false;
+	bool     output    = false;
+	uint64_t next_poll = 0;
+};
+
+MouseJoystickState g_mouse;
+
 std::size_t ControlFromName(std::string_view name) {
 	const auto info = std::find_if(CONTROL_INFO.begin(), CONTROL_INFO.end(),
 	                               [name](const auto& item) { return item.name == name; });
 	return static_cast<std::size_t>(std::distance(CONTROL_INFO.begin(), info));
 }
 
-const ControlInfo& Info(std::size_t control) {
-	EXIT_IF(control >= CONTROL_INFO.size());
-	return CONTROL_INFO[control];
-}
-
 SDL_Keycode NormalizeKey(SDL_Keycode key) {
 	switch (key) {
 		case SDLK_RSHIFT: return SDLK_LSHIFT;
@@ -103,10 +119,17 @@ bool Conflicts(const Binding& first, const Binding& second) {
 
 class InputMap {
 public:
-	InputMap(): m_custom(!Config::GetKeymap().empty()) {
+	InputMap() {
 		for (const auto& value: Config::GetKeymap()) {
 			const std::string_view entry = value;
-			const auto             split = entry.find('=');
+			if (entry.starts_with(MOUSE_SENSITIVITY)) {
+				const float sensitivity =
+				    std::strtof(value.c_str() + MOUSE_SENSITIVITY.size(), nullptr);
+				m_mouse_sensitivity =
+				    std::clamp(std::isfinite(sensitivity) ? sensitivity : 1.0f, 0.1f, 5.0f);
+				continue;
+			}
+			const auto split = entry.find('=');
 
 			Binding binding;
 			if (split != std::string_view::npos) {
@@ -118,9 +141,9 @@ public:
 				}
 			}
 
-			const bool reserved =
-			    binding.key == SDLK_ESCAPE || binding.key == SDLK_SPACE || binding.key == SDLK_F1 ||
-			    binding.key == SDLK_F11;
+			const bool reserved = binding.key == SDLK_ESCAPE || binding.key == SDLK_SPACE ||
+			                      binding.key == SDLK_F1 || binding.key == SDLK_F7 ||
+			                      binding.key == SDLK_F11;
 			if (binding.control == INVALID_CONTROL || reserved ||
 			    (binding.key == SDLK_UNKNOWN && binding.mouse_button == 0)) {
 				EXIT("Invalid input mapping: %s\n", value.c_str());
@@ -128,7 +151,8 @@ public:
 			Add(binding);
 		}
 	}
-	[[nodiscard]] bool Custom() const { return m_custom; }
+	[[nodiscard]] bool  Custom() const { return m_size != 0; }
+	[[nodiscard]] float MouseSensitivity() const { return m_mouse_sensitivity; }
 
 	[[nodiscard]] std::size_t FindKey(int key_code) const {
 		key_code = NormalizeKey(static_cast<SDL_Keycode>(key_code));
@@ -159,8 +183,8 @@ private:
 	}
 
 	std::array<Binding, CONTROL_INFO.size()> m_bindings {};
-	std::size_t                              m_size = 0;
-	bool                                     m_custom;
+	std::size_t                              m_size              = 0;
+	float                                    m_mouse_sensitivity = 1.0f;
 };
 
 const InputMap& GetInputMap() {
@@ -180,6 +204,10 @@ void SetButton(uint32_t button, bool down) {
 	}
 }
 
+void SetTouchPad(float x, bool down) {
+	Controller::ControllerTouchPad(Controller::HOST_INPUT_CONTROLLER_ID, 0, down, x, 0.5f);
+}
+
 uint32_t DefaultKeyboardButton(int key_code) {
 	switch (NormalizeKey(static_cast<SDL_Keycode>(key_code))) {
 		case SDLK_UP: return Controller::PAD_BUTTON_UP;
@@ -196,8 +224,6 @@ uint32_t DefaultKeyboardButton(int key_code) {
 		case SDLK_LCTRL: return Controller::PAD_BUTTON_R3;
 		case SDLK_RETURN:
 		case SDLK_RETURN2: return Controller::PAD_BUTTON_OPTIONS;
-		case SDLK_BACKSPACE:
-		case SDLK_TAB: return Controller::PAD_BUTTON_TOUCH_PAD;
 		default: return 0;
 	}
 }
@@ -210,7 +236,12 @@ struct StickKeys {
 };
 
 void SetStickAxis(Controller::Axis axis, bool negative, bool positive) {
-	const int value = negative == positive ? 128 : negative ? 0 : 255;
+	int value = 128;
+	if (negative && !positive) {
+		value = 0;
+	} else if (positive && !negative) {
+		value = 255;
+	}
 	Controller::ControllerAxis(Controller::HOST_INPUT_CONTROLLER_ID, axis, value);
 }
 
@@ -219,7 +250,11 @@ void SetControl(std::size_t control, bool down) {
 		return;
 	}
 
-	const auto& info = Info(control);
+	const auto& info = CONTROL_INFO[control];
+	if (info.button == Controller::PAD_BUTTON_TOUCH_PAD) {
+		SetTouchPad(info.touch_x, down);
+		return;
+	}
 	if (info.button != 0) {
 		SetButton(info.button, down);
 		return;
@@ -246,6 +281,8 @@ void DefaultKeyboardInput(int key_code, bool down) {
 	static StickKeys right;
 
 	switch (NormalizeKey(static_cast<SDL_Keycode>(key_code))) {
+		case SDLK_BACKSPACE: SetTouchPad(0.25f, down); return;
+		case SDLK_TAB: SetTouchPad(0.75f, down); return;
 		case SDLK_a:
 			left.left = down;
 			SetStickAxis(Controller::Axis::LeftX, left.left, left.right);
@@ -281,6 +318,33 @@ void DefaultKeyboardInput(int key_code, bool down) {
 		default: SetButton(DefaultKeyboardButton(key_code), down); return;
 	}
 }
+
+void MouseToJoystick(int delta_x, int delta_y) {
+	const double distance = std::hypot(delta_x, delta_y);
+	const double scale =
+	    std::clamp(distance * GetInputMap().MouseSensitivity() + 16.0, 64.0, 128.0) / distance;
+	const auto map_axis = [scale](int delta) {
+		return std::clamp(128 + static_cast<int>(std::lround(delta * scale)), 0, 255);
+	};
+	Controller::ControllerRightStick(Controller::HOST_INPUT_CONTROLLER_ID, map_axis(delta_x),
+	                                 map_axis(delta_y));
+}
+
+void CenterMouseStick() {
+	if (!g_mouse.output) {
+		return;
+	}
+	Controller::ControllerRightStick(Controller::HOST_INPUT_CONTROLLER_ID, 128, 128);
+	g_mouse.output = false;
+}
+
+bool SetRelativeMouseMode(bool enabled) {
+	if (SDL_SetRelativeMouseMode(enabled ? SDL_TRUE : SDL_FALSE) == 0) {
+		return true;
+	}
+	LOGF("Mouse-to-joystick relative mode failed: %s\n", SDL_GetError());
+	return false;
+}
 } // namespace
 
 void HostInputInit() {
@@ -303,4 +367,63 @@ void HostInputMouseButton(uint8_t mouse_button, bool down) {
 	}
 }
 
+void HostInputToggleMouseToJoystick() {
+	if (g_mouse.enabled) {
+		SetRelativeMouseMode(false);
+		CenterMouseStick();
+		g_mouse = {};
+		LOGF("Mouse to right stick: disabled\n");
+		return;
+	}
+
+	if (!SetRelativeMouseMode(true)) {
+		return;
+	}
+	int ignored_x = 0;
+	int ignored_y = 0;
+	SDL_GetRelativeMouseState(&ignored_x, &ignored_y);
+	g_mouse.enabled   = true;
+	g_mouse.next_poll = SDL_GetTicks64() + MOUSE_POLL_INTERVAL_MS;
+	LOGF("Mouse to right stick: enabled (F7 to release)\n");
+}
+
+int PollMouse(uint64_t now_ms) {
+	if (now_ms < g_mouse.next_poll) {
+		return static_cast<int>(g_mouse.next_poll - now_ms);
+	}
+	g_mouse.next_poll = now_ms + MOUSE_POLL_INTERVAL_MS;
+
+	int delta_x = 0;
+	int delta_y = 0;
+	SDL_GetRelativeMouseState(&delta_x, &delta_y);
+	if (delta_x == 0 && delta_y == 0) {
+		CenterMouseStick();
+		return MOUSE_POLL_INTERVAL_MS;
+	}
+
+	MouseToJoystick(delta_x, delta_y);
+	g_mouse.output = true;
+	return MOUSE_POLL_INTERVAL_MS;
+}
+
+bool HostInputWaitEvent(SDL_Event* event) {
+	if (!g_mouse.enabled || SDL_GetKeyboardFocus() == nullptr) {
+		CenterMouseStick();
+		if (SDL_WaitEvent(event) == 0) {
+			EXIT("%s\n", SDL_GetError());
+		}
+		return true;
+	}
+
+	const int timeout_ms = PollMouse(SDL_GetTicks64());
+	SDL_ClearError();
+	if (SDL_WaitEventTimeout(event, timeout_ms) != 0) {
+		return true;
+	}
+	if (SDL_GetError()[0] != '\0') {
+		EXIT("%s\n", SDL_GetError());
+	}
+	return false;
+}
+
 } // namespace Libs::Graphics

src/graphics/presentation/window/hostInput.h

@@ -3,11 +3,15 @@
 
 #include <cstdint>
 
+union SDL_Event;
+
 namespace Libs::Graphics {
 
-void HostInputInit();
-void HostInputKey(int key_code, bool down);
-void HostInputMouseButton(uint8_t mouse_button, bool down);
+void               HostInputInit();
+void               HostInputKey(int key_code, bool down);
+void               HostInputMouseButton(uint8_t mouse_button, bool down);
+void               HostInputToggleMouseToJoystick();
+[[nodiscard]] bool HostInputWaitEvent(SDL_Event* event);
 
 } // namespace Libs::Graphics
 

src/graphics/presentation/window/swapchain.cpp

@@ -1,30 +1,9 @@
-#include "SDL.h"
-#include "SDL_error.h"
-#include "SDL_events.h"
-#include "SDL_gamecontroller.h"
-#include "SDL_hints.h"
-#include "SDL_joystick.h"
-#include "SDL_keyboard.h"
-#include "SDL_keycode.h"
-#include "SDL_mouse.h"
-#include "SDL_pixels.h"
-#include "SDL_rwops.h"
-#include "SDL_stdinc.h"
-#include "SDL_surface.h"
-#include "SDL_thread.h"
-#include "SDL_touch.h"
-#include "SDL_video.h"
-#include "SDL_vulkan.h"
 #include "common/assert.h"
 #include "common/common.h"
 #include "common/emulatorConfig.h"
-#include "common/file.h"
 #include "common/logging/log.h"
 #include "common/profiler.h"
-#include "common/stringUtils.h"
-#include "common/systemInfo.h"
 #include "common/threads.h"
-#include "common/timer.h"
 #include "graphics/host_gpu/graphicContext.h"
 #include "graphics/host_gpu/renderer/render.h"
 #include "graphics/host_gpu/renderer/renderContext.h"
@@ -34,19 +13,11 @@
 #include "graphics/presentation/presenter.h"
 #include "graphics/presentation/videoOut.h"
 #include "graphics/presentation/window/windowInternal.h"
-#include "libs/controller.h"
-#include "loader/systemContent.h"
 
 #include <algorithm>
-#include <array>
-#include <cstdio>
-#include <cstdlib>
-#include <cstring>
 #include <deque>
-#include <fmt/format.h>
 #include <limits>
 #include <memory>
-#include <string>
 #include <vector>
 #include <vulkan/vk_platform.h>
 
@@ -58,10 +29,10 @@
 namespace Libs::Graphics {
 
 struct Presenter::Frame {
-	VulkanImage                    image;
-	std::unique_ptr<CommandBuffer> present_commands;
-	bool                           busy         = false;
-	bool                           reusing_last = false;
+	VulkanImage image;
+	uint64_t    present_tick = 0;
+	bool        busy         = false;
+	bool        reusing_last = false;
 
 	void Configure(GraphicContext& graphics, vk::Extent2D extent, vk::Format format);
 	void Transit(vk::CommandBuffer command, vk::ImageLayout layout, vk::AccessFlags2 access);
@@ -71,12 +42,11 @@ struct Presenter::Frame {
 
 class FramePool final {
 public:
-	explicit FramePool(WindowContext& window): m_window(window) {}
+	FramePool(WindowContext& window, CommandScheduler& scheduler)
+	    : m_window(window), m_scheduler(scheduler) {}
 	~FramePool() {
+		m_scheduler.Wait(m_scheduler.CurrentTick() - 1);
 		for (auto& frame: m_frames) {
-			if (frame->present_commands != nullptr) {
-				frame->present_commands->WaitForFenceOnly();
-			}
 			if (frame->image.image != nullptr) {
 				m_window.graphic_ctx.DeleteImage(frame->image);
 			}
@@ -188,15 +158,10 @@ public:
 	}
 
 private:
-	static void WaitForFrame(Presenter::Frame& frame) {
-		// The producer only waits here. Reset stays on the presentation thread that owns the
-		// allocating Vulkan command pool.
-		if (frame.present_commands != nullptr) {
-			frame.present_commands->WaitForFenceOnly();
-		}
-	}
+	void WaitForFrame(Presenter::Frame& frame) { m_scheduler.Wait(frame.present_tick); }
 
 	WindowContext&                                 m_window;
+	CommandScheduler&                              m_scheduler;
 	Common::Mutex                                  m_mutex;
 	Common::CondVar                                m_available;
 	std::vector<std::unique_ptr<Presenter::Frame>> m_frames;
@@ -330,7 +295,7 @@ public:
 	[[nodiscard]] Status AcquireNextImage();
 	[[nodiscard]] bool   PrepareImeOverlay();
 	void RecordPresentCommands(CommandBuffer& command, VulkanImage& source, bool draw_ime_overlay);
-	void Submit(CommandBuffer& command);
+	uint64_t             Submit(CommandScheduler& scheduler);
 	[[nodiscard]] Status Present();
 
 	[[nodiscard]] uint32_t ImageCount() const noexcept {
@@ -340,7 +305,6 @@ public:
 
 private:
 	void Destroy();
-	void RefreshSurfaceSize();
 
 	WindowContext&              m_window;
 	vk::SwapchainKHR            m_handle = nullptr;
@@ -358,7 +322,7 @@ private:
 struct Presenter::Impl {
 	explicit Impl(WindowContext& owner)
 	    : renderer(*owner.render_context), window(owner), swapchain(owner),
-	      present_scheduler(renderer, owner.graphic_ctx), frames(owner) {
+	      present_scheduler(renderer, owner.graphic_ctx), frames(owner, present_scheduler) {
 		EXIT_IF(owner.render_context == nullptr);
 		swapchain.Create();
 		frames.Initialize(swapchain.ImageCount(), swapchain.Format());
@@ -402,12 +366,12 @@ struct Presenter::Impl {
 
 void Swapchain::Create() {
 	auto& graphics = m_window.graphic_ctx;
-	EXIT_IF(graphics.screen_width == 0);
-	EXIT_IF(graphics.screen_height == 0);
 	EXIT_IF(graphics.device == nullptr);
 	EXIT_IF(m_window.surface == nullptr);
 
 	Common::LockGuard lock(m_window.mutex);
+	EXIT_IF(graphics.screen_width == 0);
+	EXIT_IF(graphics.screen_height == 0);
 	const auto&       surface = m_window.surface_capabilities;
 	EXIT_NOT_IMPLEMENTED(surface.formats.empty());
 
@@ -465,7 +429,21 @@ void Swapchain::Create() {
 	create_info.imageSharingMode = vk::SharingMode::eExclusive;
 	create_info.preTransform     = transform;
 	create_info.compositeAlpha   = composite;
-	create_info.presentMode      = vk::PresentModeKHR::eFifo;
+	switch (Config::GetPresentMode()) {
+		case Config::PresentMode::Mailbox:
+			create_info.presentMode = vk::PresentModeKHR::eMailbox;
+			break;
+		case Config::PresentMode::Immediate:
+			create_info.presentMode = vk::PresentModeKHR::eImmediate;
+			break;
+		case Config::PresentMode::Fifo:
+		default: create_info.presentMode = vk::PresentModeKHR::eFifo; break;
+	}
+	if (std::find(surface.present_modes.begin(), surface.present_modes.end(),
+	              create_info.presentMode) == surface.present_modes.end()) {
+		LOGF("warning: requested present mode is unavailable; falling back to Fifo\n");
+		create_info.presentMode = vk::PresentModeKHR::eFifo;
+	}
 	create_info.clipped          = VK_TRUE;
 	RequireVulkanSuccess(graphics.device.createSwapchainKHR(&create_info, nullptr, &m_handle),
 	                     "vkCreateSwapchainKHR");
@@ -560,30 +538,18 @@ void Swapchain::Destroy() {
 	m_render_complete.clear();
 }
 
-void Swapchain::RefreshSurfaceSize() {
-	int width  = 0;
-	int height = 0;
-	SDL_Vulkan_GetDrawableSize(m_window.window, &width, &height);
-	if (width > 0 && height > 0) {
-		m_window.graphic_ctx.screen_width  = static_cast<uint32_t>(width);
-		m_window.graphic_ctx.screen_height = static_cast<uint32_t>(height);
-	}
-
-	m_window.RefreshSurfaceCapabilities();
-}
-
 void Swapchain::Recreate(bool surface_lost) {
 	Destroy();
 	if (surface_lost) {
 #if defined(__APPLE__)
 		// Surface recreation goes through SDL_Vulkan_CreateSurface, which touches the
 		// window's view/layer and must run on the main thread on macOS.
-		m_window.RunOnMainThread([this] { m_window.RecreateSurface(); }, true);
+		m_window.RunOnMainThread([this] { m_window.RecreateSurface(); });
 #else
 		m_window.RecreateSurface();
 #endif
 	}
-	RefreshSurfaceSize();
+	m_window.RefreshSurfaceCapabilities();
 	Create();
 }
 
@@ -628,7 +594,6 @@ void Swapchain::RecordPresentCommands(CommandBuffer& command, VulkanImage& sourc
 	}
 	EXIT_IF(m_image_index >= m_images.size());
 	auto vk_command = command.Handle();
-	command.Begin();
 
 	vk::ImageMemoryBarrier to_transfer {};
 	to_transfer.sType                           = vk::StructureType::eImageMemoryBarrier;
@@ -699,15 +664,14 @@ void Swapchain::RecordPresentCommands(CommandBuffer& command, VulkanImage& sourc
 		                           vk::DependencyFlagBits::eByRegion, 0, nullptr, 0, nullptr, 1,
 		                           &to_present);
 	}
-	command.End();
 }
 
-void Swapchain::Submit(CommandBuffer& command) {
+uint64_t Swapchain::Submit(CommandScheduler& scheduler) {
 	EXIT_IF(m_frame_index >= m_image_acquired.size() || m_image_index >= m_render_complete.size());
 	SubmitInfo submit;
 	submit.AddWait(m_image_acquired[m_frame_index], 1, vk::PipelineStageFlagBits::eTransfer);
 	submit.AddSignal(m_render_complete[m_image_index]);
-	command.Execute(submit);
+	return scheduler.Submit(submit);
 }
 
 Swapchain::Status Swapchain::Present() {
@@ -782,15 +746,9 @@ Presenter::Frame& Presenter::PrepareBlankFrame(uint32_t width, uint32_t height,
 		EXIT_IF(producer->IsInvalid());
 		frame->Clear(*producer, clear);
 	} else {
-		if (frame->present_commands == nullptr) {
-			frame->present_commands = std::make_unique<CommandBuffer>(m_impl->present_scheduler);
-		}
-		auto& command = *frame->present_commands;
-		command.WaitForFenceAndReset();
-		command.Begin();
+		auto& command = m_impl->present_scheduler.BeginCommand();
 		frame->Clear(command, clear);
-		command.End();
-		command.Execute();
+		frame->present_tick = m_impl->present_scheduler.Submit();
 	}
 	return *frame;
 }
@@ -817,29 +775,6 @@ void Presenter::Present(Frame& frame, bool reuse) {
 	KYTY_PROFILER_FUNCTION();
 	m_impl->frames.ValidateForPresent(&frame, reuse);
 
-	auto& window = m_impl->window;
-	if (window.window_hidden) {
-#if defined(__APPLE__)
-		// AppKit traps if a window is shown off the main thread; marshal and wait so the
-		// swapchain below is recreated against a visible window.
-		window.RunOnMainThread(
-		    [&window] {
-			    window.UpdateIcon();
-			    SDL_ShowWindow(window.window);
-			    SDL_RaiseWindow(window.window);
-		    },
-		    true);
-#else
-		window.UpdateIcon();
-
-		SDL_ShowWindow(window.window);
-		SDL_RaiseWindow(window.window);
-#endif
-
-		window.window_hidden = false;
-		m_impl->RecoverSwapchain(Swapchain::Status::Recreate);
-	}
-
 	const auto ime_visual = GetImeVisualState();
 	auto&      swapchain  = m_impl->swapchain;
 	for (uint32_t attempt = 0; attempt < 2; attempt++) {
@@ -848,16 +783,12 @@ void Presenter::Present(Frame& frame, bool reuse) {
 			m_impl->RecoverSwapchain(status);
 			continue;
 		}
-		if (frame.present_commands == nullptr) {
-			frame.present_commands = std::make_unique<CommandBuffer>(m_impl->present_scheduler);
-		}
 		{
 			Common::LockGuard render_lock(m_impl->renderer.GetMutex());
-			frame.present_commands->WaitForFenceAndReset();
-			auto&      command          = *frame.present_commands;
-			const bool draw_ime_overlay = ime_visual.active && swapchain.PrepareImeOverlay();
+			auto&             command          = m_impl->present_scheduler.BeginCommand();
+			const bool        draw_ime_overlay = ime_visual.active && swapchain.PrepareImeOverlay();
 			swapchain.RecordPresentCommands(command, frame.image, draw_ime_overlay);
-			swapchain.Submit(command);
+			frame.present_tick = swapchain.Submit(m_impl->present_scheduler);
 		}
 		status = swapchain.Present();
 		if (status != Swapchain::Status::Success) {
@@ -866,7 +797,7 @@ void Presenter::Present(Frame& frame, bool reuse) {
 		}
 
 		m_impl->presented_ime_revision.store(ime_visual.revision, std::memory_order_release);
-		window.UpdateTitle();
+		m_impl->window.UpdateTitle();
 		m_impl->frames.Release(&frame, true);
 		return;
 	}

src/graphics/presentation/window/vulkanWindow.cpp

@@ -22,7 +22,6 @@
 #include "common/logging/log.h"
 #include "common/profiler.h"
 #include "common/stringUtils.h"
-#include "common/systemInfo.h"
 #include "common/threads.h"
 #include "common/timer.h"
 #include "graphics/host_gpu/graphicContext.h"
@@ -158,6 +157,10 @@ static void VulkanFindPhysicalDevice(vk::Instance instance, vk::SurfaceKHR surfa
 	    });
 	EXIT_NOT_IMPLEMENTED(devices.empty());
 
+	if (Config::GetGpuIndex() >= 0) {
+		devices = {devices[Config::GetGpuIndex()]};
+	}
+
 	vk::PhysicalDevice  best_device       = nullptr;
 	uint32_t            best_queue_family = static_cast<uint32_t>(-1);
 	SurfaceCapabilities best_capabilities;
@@ -199,7 +202,15 @@ static void VulkanFindPhysicalDevice(vk::Instance instance, vk::SurfaceKHR surfa
 
 		vk::PhysicalDeviceVulkan12Features features12 {};
 		features12.sType = vk::StructureType::ePhysicalDeviceVulkan12Features;
+#if defined(__APPLE__)
 		features12.pNext = &depth_clip_control;
+#else
+		vk::PhysicalDeviceFragmentShaderBarycentricFeaturesKHR fragment_barycentric {};
+		fragment_barycentric.sType =
+		    vk::StructureType::ePhysicalDeviceFragmentShaderBarycentricFeaturesKHR;
+		fragment_barycentric.pNext = &depth_clip_control;
+		features12.pNext           = &fragment_barycentric;
+#endif
 		features13.pNext = &features12;
 
 		device_features2.sType = vk::StructureType::ePhysicalDeviceFeatures2;
@@ -231,6 +242,12 @@ static void VulkanFindPhysicalDevice(vk::Instance instance, vk::SurfaceKHR surfa
 			skip_device = true;
 #endif
 		}
+#if !defined(__APPLE__)
+		if (fragment_barycentric.fragmentShaderBarycentric != VK_TRUE) {
+			LOGF("fragmentShaderBarycentric is not supported\n");
+			skip_device = true;
+		}
+#endif
 
 		if (features12.samplerMirrorClampToEdge != VK_TRUE) {
 			LOGF("samplerMirrorClampToEdge is not supported\n");
@@ -262,6 +279,22 @@ static void VulkanFindPhysicalDevice(vk::Instance instance, vk::SurfaceKHR surfa
 			LOGF("depthBiasClamp is not supported\n");
 			skip_device = true;
 		}
+		if (device_features2.features.shaderClipDistance != VK_TRUE) {
+			LOGF("shaderClipDistance is not supported\n");
+			skip_device = true;
+		}
+		if (device_features2.features.shaderCullDistance != VK_TRUE) {
+			LOGF("shaderCullDistance is not supported\n");
+			skip_device = true;
+		}
+		if (device_features2.features.largePoints != VK_TRUE) {
+			LOGF("largePoints is not supported\n");
+			skip_device = true;
+		}
+		if (features12.shaderOutputLayer != VK_TRUE) {
+			LOGF("shaderOutputLayer is not supported\n");
+			skip_device = true;
+		}
 
 		if (device_features2.features.fragmentStoresAndAtomics != VK_TRUE) {
 			LOGF("fragmentStoresAndAtomics is not supported\n");
@@ -463,14 +496,19 @@ static void VulkanInitSubgroupSizeControl(vk::PhysicalDevice physical_device,
 	graphics.min_subgroup_size             = subgroup_size_control.minSubgroupSize;
 	graphics.max_subgroup_size             = subgroup_size_control.maxSubgroupSize;
 	graphics.required_subgroup_size_stages = subgroup_size_control.requiredSubgroupSizeStages;
-	graphics.subgroup_size_control_enabled = features13.subgroupSizeControl == VK_TRUE &&
-	                                         subgroup_size_control.minSubgroupSize <= 64 &&
-	                                         subgroup_size_control.maxSubgroupSize >= 64;
+	graphics.compute_subgroup_size_control_enabled =
+	    features13.subgroupSizeControl == VK_TRUE &&
+	    (graphics.required_subgroup_size_stages & vk::ShaderStageFlagBits::eCompute) &&
+	    subgroup_size_control.minSubgroupSize <= 64 &&
+	    subgroup_size_control.maxSubgroupSize >= 64;
+	graphics.compute_wave64_supported =
+	    graphics.subgroup_size == 64u || graphics.compute_subgroup_size_control_enabled;
 
-	LOGF("Vulkan subgroup: default=%u min=%u max=%u stages=0x%08x size_control=%s\n",
+	LOGF("Vulkan subgroup: default=%u min=%u max=%u stages=0x%08x size_control=%s wave64=%s\n",
 	     graphics.subgroup_size, graphics.min_subgroup_size, graphics.max_subgroup_size,
 	     static_cast<vk::ShaderStageFlags::MaskType>(graphics.required_subgroup_size_stages),
-	     graphics.subgroup_size_control_enabled ? "true" : "false");
+	     graphics.compute_subgroup_size_control_enabled ? "true" : "false",
+	     graphics.compute_wave64_supported ? "true" : "false");
 }
 
 static vk::Device VulkanCreateDevice(vk::PhysicalDevice physical_device, const VulkanExtensions& r,
@@ -514,10 +552,6 @@ static vk::Device VulkanCreateDevice(vk::PhysicalDevice physical_device, const V
 	features12.pNext                    = &depth_clip_control;
 	features12.samplerMirrorClampToEdge = VK_TRUE;
 
-	vk::PhysicalDeviceSubgroupSizeControlFeatures subgroup_size_control {};
-	subgroup_size_control.sType = vk::StructureType::ePhysicalDeviceSubgroupSizeControlFeatures;
-	subgroup_size_control.pNext = &features12;
-
 	vk::PhysicalDeviceVulkan13Features supported_features13 {};
 	supported_features13.sType = vk::StructureType::ePhysicalDeviceVulkan13Features;
 
@@ -526,6 +560,14 @@ static vk::Device VulkanCreateDevice(vk::PhysicalDevice physical_device, const V
 	supported_features12.pNext = nullptr;
 	supported_features13.pNext = &supported_features12;
 
+#if !defined(__APPLE__)
+	vk::PhysicalDeviceFragmentShaderBarycentricFeaturesKHR supported_fragment_barycentric {};
+	supported_fragment_barycentric.sType =
+	    vk::StructureType::ePhysicalDeviceFragmentShaderBarycentricFeaturesKHR;
+	supported_fragment_barycentric.pNext = nullptr;
+	supported_features12.pNext           = &supported_fragment_barycentric;
+#endif
+
 	const auto robustness2_ext_enabled =
 	    HasExtension(device_extensions, VK_EXT_ROBUSTNESS_2_EXTENSION_NAME);
 
@@ -533,7 +575,11 @@ static vk::Device VulkanCreateDevice(vk::PhysicalDevice physical_device, const V
 	supported_robustness2.sType = vk::StructureType::ePhysicalDeviceRobustness2FeaturesEXT;
 	supported_robustness2.pNext = nullptr;
 	if (robustness2_ext_enabled) {
+#if defined(__APPLE__)
 		supported_features12.pNext = &supported_robustness2;
+#else
+		supported_fragment_barycentric.pNext = &supported_robustness2;
+#endif
 	}
 
 	vk::PhysicalDeviceFeatures2 supported_features2 {};
@@ -548,15 +594,19 @@ static vk::Device VulkanCreateDevice(vk::PhysicalDevice physical_device, const V
 	                     supported_features13.synchronization2 != VK_TRUE);
 	EXIT_NOT_IMPLEMENTED(supported_features2.features.sampleRateShading != VK_TRUE);
 	EXIT_NOT_IMPLEMENTED(supported_features2.features.depthBiasClamp != VK_TRUE);
+	EXIT_NOT_IMPLEMENTED(supported_features2.features.shaderClipDistance != VK_TRUE);
+	EXIT_NOT_IMPLEMENTED(supported_features2.features.shaderCullDistance != VK_TRUE);
+	EXIT_NOT_IMPLEMENTED(supported_features2.features.largePoints != VK_TRUE);
+	EXIT_NOT_IMPLEMENTED(supported_features2.features.shaderInt64 != VK_TRUE);
+	EXIT_NOT_IMPLEMENTED(supported_features2.features.vertexPipelineStoresAndAtomics != VK_TRUE);
+	EXIT_NOT_IMPLEMENTED(supported_features12.shaderOutputLayer != VK_TRUE);
+	EXIT_NOT_IMPLEMENTED(supported_features12.bufferDeviceAddress != VK_TRUE);
+#if !defined(__APPLE__)
+	EXIT_NOT_IMPLEMENTED(supported_fragment_barycentric.fragmentShaderBarycentric != VK_TRUE);
+#endif
 	features12.timelineSemaphore = VK_TRUE;
-
-	if (Config::GpuAssistedValidationEnabled()) {
-		if (supported_features12.bufferDeviceAddress != VK_TRUE) {
-			EXIT("--gpu-assisted-validation requires bufferDeviceAddress, unsupported by this "
-			     "device\n");
-		}
-		features12.bufferDeviceAddress = VK_TRUE;
-	}
+	features12.shaderOutputLayer = VK_TRUE;
+	features12.bufferDeviceAddress = VK_TRUE;
 
 	vk::PhysicalDeviceFeatures device_features {};
 	device_features.fragmentStoresAndAtomics = VK_TRUE;
@@ -571,35 +621,45 @@ static vk::Device VulkanCreateDevice(vk::PhysicalDevice physical_device, const V
 	device_features.tessellationShader                   = VK_TRUE;
 	device_features.sampleRateShading                    = VK_TRUE;
 	device_features.depthBiasClamp                       = VK_TRUE;
+	device_features.shaderClipDistance                   = VK_TRUE;
+	device_features.shaderCullDistance                   = VK_TRUE;
+	device_features.largePoints                          = VK_TRUE;
+	device_features.vertexPipelineStoresAndAtomics       = VK_TRUE;
 	graphics.sample_rate_shading_enabled                 = true;
-	device_features.vertexPipelineStoresAndAtomics =
-	    supported_features2.features.vertexPipelineStoresAndAtomics;
-	if (Config::GpuAssistedValidationEnabled()) {
-		device_features.shaderInt64 = supported_features2.features.shaderInt64;
-	}
-
-	if (graphics.subgroup_size_control_enabled &&
-	    supported_features13.subgroupSizeControl == VK_TRUE) {
-		subgroup_size_control.subgroupSizeControl = VK_TRUE;
-	}
-
-	const auto* base_feature_chain = (subgroup_size_control.subgroupSizeControl == VK_TRUE
-	                                      ? static_cast<const void*>(&subgroup_size_control)
-	                                      : static_cast<const void*>(&features12));
+	device_features.shaderInt64 = VK_TRUE;
 
 	vk::PhysicalDeviceRobustness2FeaturesEXT robustness2 {};
 	robustness2.sType = vk::StructureType::ePhysicalDeviceRobustness2FeaturesEXT;
-	robustness2.pNext = const_cast<void*>(base_feature_chain);
+#if defined(__APPLE__)
+	robustness2.pNext = &features12;
+#else
+	vk::PhysicalDeviceFragmentShaderBarycentricFeaturesKHR fragment_barycentric {};
+	fragment_barycentric.sType =
+	    vk::StructureType::ePhysicalDeviceFragmentShaderBarycentricFeaturesKHR;
+	fragment_barycentric.pNext                     = &features12;
+	fragment_barycentric.fragmentShaderBarycentric = VK_TRUE;
+	robustness2.pNext                              = &fragment_barycentric;
+#endif
 	if (robustness2_ext_enabled) {
 		robustness2.robustBufferAccess2 = supported_robustness2.robustBufferAccess2;
 		robustness2.robustImageAccess2  = supported_robustness2.robustImageAccess2;
 		robustness2.nullDescriptor      = supported_robustness2.nullDescriptor;
 	}
 
+	const bool subgroup_size_control_enabled =
+	    graphics.compute_subgroup_size_control_enabled &&
+	    supported_features13.subgroupSizeControl == VK_TRUE;
+
 	auto features13 = required_features13;
-	features13.pNext =
-	    robustness2_ext_enabled ? &robustness2 : const_cast<void*>(base_feature_chain);
-	features13.robustImageAccess = supported_features13.robustImageAccess;
+#if defined(__APPLE__)
+	features13.pNext = robustness2_ext_enabled ? static_cast<void*>(&robustness2)
+	                                           : static_cast<void*>(&features12);
+#else
+	features13.pNext = robustness2_ext_enabled ? static_cast<void*>(&robustness2)
+	                                           : static_cast<void*>(&fragment_barycentric);
+#endif
+	features13.robustImageAccess   = supported_features13.robustImageAccess;
+	features13.subgroupSizeControl = subgroup_size_control_enabled ? VK_TRUE : VK_FALSE;
 
 	LOGF("Vulkan robustness: robustImageAccess=%s robustImageAccess2=%s\n",
 	     features13.robustImageAccess == VK_TRUE ? "true" : "false",
@@ -843,7 +903,7 @@ void WindowContext::CreateVulkan() {
 		EXIT("--spirv-debug-printf and --gpu-assisted-validation are mutually exclusive\n");
 	}
 
-	vk::ValidationFeatureEnableEXT enabled_features[3] = {};
+	vk::ValidationFeatureEnableEXT enabled_features[3]    = {};
 	uint32_t                       enabled_features_count = 0;
 #ifdef KYTY_ENABLE_BEST_PRACTICES
 	enabled_features[enabled_features_count++] = vk::ValidationFeatureEnableEXT::eBestPractices;
@@ -886,6 +946,18 @@ void WindowContext::CreateVulkan() {
 	inst_info.sType                   = vk::StructureType::eInstanceCreateInfo;
 	inst_info.pNext                   = (r.enable_validation_layers ? &dbg_create_info : nullptr);
 	inst_info.flags                   = {};
+#if defined(__APPLE__)
+	// MoltenVK requires VK_KHR_portability_enumeration + flag to surface
+	// portability devices. Without this, enumeratePhysicalDevices hides the
+	// MoltenVK adapter on some driver versions.
+	if (HasExtension(r.available_extensions, VK_KHR_PORTABILITY_ENUMERATION_EXTENSION_NAME)) {
+		if (!HasExtension(r.required_extensions, VK_KHR_PORTABILITY_ENUMERATION_EXTENSION_NAME)) {
+			r.required_extensions.push_back(VK_KHR_PORTABILITY_ENUMERATION_EXTENSION_NAME);
+		}
+		inst_info.flags |= vk::InstanceCreateFlagBits::eEnumeratePortabilityKHR;
+		LOGF("Vulkan instance: enabled %s\n", VK_KHR_PORTABILITY_ENUMERATION_EXTENSION_NAME);
+	}
+#endif
 	inst_info.pApplicationInfo        = &app_info;
 	inst_info.enabledExtensionCount   = static_cast<uint32_t>(r.required_extensions.size());
 	inst_info.ppEnabledExtensionNames = r.required_extensions.data();
@@ -931,6 +1003,7 @@ void WindowContext::CreateVulkan() {
 #else
 	device_extensions.push_back(VK_EXT_DEPTH_CLIP_ENABLE_EXTENSION_NAME);
 	device_extensions.push_back(VK_EXT_COLOR_WRITE_ENABLE_EXTENSION_NAME);
+	device_extensions.push_back(VK_KHR_FRAGMENT_SHADER_BARYCENTRIC_EXTENSION_NAME);
 #endif
 
 #ifdef KYTY_ENABLE_DEBUG_PRINTF
@@ -948,7 +1021,15 @@ void WindowContext::CreateVulkan() {
 		EXIT("Could not find suitable device");
 	}
 
-	graphic_ctx.physical_device.getProperties(&graphic_ctx.physical_device_properties);
+	vk::PhysicalDevicePushDescriptorProperties push_descriptor_properties {};
+	vk::PhysicalDeviceProperties2              physical_device_properties {};
+	push_descriptor_properties.sType = vk::StructureType::ePhysicalDevicePushDescriptorProperties;
+	push_descriptor_properties.pNext = nullptr;
+	physical_device_properties.sType = vk::StructureType::ePhysicalDeviceProperties2;
+	physical_device_properties.pNext = &push_descriptor_properties;
+	graphic_ctx.physical_device.getProperties2(&physical_device_properties);
+	graphic_ctx.physical_device_properties = physical_device_properties.properties;
+	graphic_ctx.max_push_descriptors       = push_descriptor_properties.maxPushDescriptors;
 	graphic_ctx.physical_device.getMemoryProperties(&graphic_ctx.physical_device_memory_properties);
 	const auto& device_properties = graphic_ctx.GetPhysicalDeviceProperties();
 
@@ -971,10 +1052,6 @@ void WindowContext::CreateVulkan() {
 		}
 	}
 
-	memcpy(device_name, device_properties.deviceName, sizeof(device_name));
-	std::snprintf(processor_name, sizeof(processor_name), "%s",
-	              Common::GetSystemInfo().ProcessorName.c_str());
-
 	VulkanInitSubgroupSizeControl(graphic_ctx.physical_device, graphic_ctx);
 
 	graphic_ctx.device = VulkanCreateDevice(graphic_ctx.physical_device, r, queue_family,
@@ -1013,7 +1090,6 @@ void WindowContext::RecreateSurface() {
 		EXIT("Could not recreate the Vulkan surface: %s\n", SDL_GetError());
 	}
 	surface = native_surface;
-	RefreshSurfaceCapabilities();
 }
 
 WindowContext::~WindowContext() {

src/graphics/presentation/window/window.cpp

@@ -3,6 +3,7 @@
 #include "SDL.h"
 #include "SDL_error.h"
 #include "SDL_events.h"
+#include "SDL_filesystem.h"
 #include "SDL_gamecontroller.h"
 #include "SDL_hints.h"
 #include "SDL_joystick.h"
@@ -23,12 +24,12 @@
 #include "common/file.h"
 #include "common/logging/log.h"
 #include "common/profiler.h"
-#include "common/stringUtils.h"
 #include "common/systemInfo.h"
 #include "common/threads.h"
 #include "common/timer.h"
 #include "graphics/host_gpu/graphicContext.h"
 #include "graphics/host_gpu/renderer/render.h"
+#include "graphics/host_gpu/renderer/renderContext.h"
 #include "graphics/host_gpu/vma.h"
 #include "graphics/host_gpu/vulkanCommon.h"
 #include "graphics/presentation/imeOverlay.h"
@@ -39,10 +40,8 @@
 #include "libs/controller.h"
 #include "loader/systemContent.h"
 
-#include <algorithm>
-#include <cstdio>
 #include <cstdlib>
-#include <cstring>
+#include <fmt/format.h>
 #include <memory>
 #include <string>
 #include <vector>
@@ -52,8 +51,6 @@
 #define STBI_NO_SIMD
 #include "stb_image.h"
 
-#include <fmt/format.h>
-
 // IWYU pragma: no_include <intrin.h>
 
 #define KYTY_ENABLE_DEBUG_PRINTF
@@ -79,7 +76,6 @@ static uint32_t ControllerButtonToPadButton(int button) {
 		case SDL_CONTROLLER_BUTTON_B: return Controller::PAD_BUTTON_CIRCLE;
 		case SDL_CONTROLLER_BUTTON_X: return Controller::PAD_BUTTON_SQUARE;
 		case SDL_CONTROLLER_BUTTON_Y: return Controller::PAD_BUTTON_TRIANGLE;
-		case SDL_CONTROLLER_BUTTON_BACK: return Controller::PAD_BUTTON_TOUCH_PAD;
 		case SDL_CONTROLLER_BUTTON_START: return Controller::PAD_BUTTON_OPTIONS;
 		case SDL_CONTROLLER_BUTTON_LEFTSTICK: return Controller::PAD_BUTTON_L3;
 		case SDL_CONTROLLER_BUTTON_RIGHTSTICK: return Controller::PAD_BUTTON_R3;
@@ -89,6 +85,7 @@ static uint32_t ControllerButtonToPadButton(int button) {
 		case SDL_CONTROLLER_BUTTON_DPAD_DOWN: return Controller::PAD_BUTTON_DOWN;
 		case SDL_CONTROLLER_BUTTON_DPAD_LEFT: return Controller::PAD_BUTTON_LEFT;
 		case SDL_CONTROLLER_BUTTON_DPAD_RIGHT: return Controller::PAD_BUTTON_RIGHT;
+		case SDL_CONTROLLER_BUTTON_TOUCHPAD: return Controller::PAD_BUTTON_TOUCH_PAD;
 		default: return 0;
 	}
 }
@@ -197,8 +194,6 @@ std::unique_ptr<WindowContext> g_window;
 } // namespace
 
 constexpr const char* KYTY_SDL_WINDOW_CAPTION = "Game";
-constexpr uint32_t    KYTY_SDL_WINDOW_FLAGS =
-    (static_cast<uint32_t>(SDL_WINDOW_HIDDEN) | static_cast<uint32_t>(SDL_WINDOW_VULKAN));
 constexpr int KYTY_SDL_WINDOWPOS_CENTERED = SDL_WINDOWPOS_CENTERED; /*NOLINT(hicpp-signed-bitwise)*/
 
 static void SetPause(WindowLoopState& game, bool flag) {
@@ -407,6 +402,7 @@ static void GameEventDidEnterForeground(WindowLoopState& game) {
 
 void WindowContext::Resize(uint32_t new_width, uint32_t new_height) {
 	EXIT_IF(new_width == 0 || new_height == 0);
+	Common::LockGuard lock(mutex);
 	graphic_ctx.screen_width  = new_width;
 	graphic_ctx.screen_height = new_height;
 }
@@ -511,6 +507,13 @@ void WindowContext::ProcessEvent(double time_s) {
 	auto& game  = loop;
 	auto* event = &game.event;
 	EXIT_IF(SDL_GetEventState(SDL_DISPLAYEVENT) != SDL_ENABLE);
+	if ((event->type == SDL_KEYDOWN || event->type == SDL_KEYUP) &&
+	    event->key.keysym.sym == SDLK_F7) {
+		if (event->type == SDL_KEYDOWN && event->key.repeat == 0) {
+			HostInputToggleMouseToJoystick();
+		}
+		return;
+	}
 	if (ProcessImeInput(*event)) {
 		return;
 	}
@@ -702,6 +705,16 @@ void WindowContext::ProcessEvent(double time_s) {
 			break;
 		}
 
+		case SDL_CONTROLLERTOUCHPADDOWN:
+		case SDL_CONTROLLERTOUCHPADMOTION:
+		case SDL_CONTROLLERTOUCHPADUP:
+			if (event->ctouchpad.touchpad == 0) {
+				Controller::ControllerTouchPad(event->ctouchpad.which, event->ctouchpad.finger,
+				                               event->ctouchpad.type != SDL_CONTROLLERTOUCHPADUP,
+				                               event->ctouchpad.x, event->ctouchpad.y);
+			}
+			break;
+
 		case SDL_CONTROLLERDEVICEADDED:
 		case SDL_CONTROLLERDEVICEREMOVED:
 		case SDL_CONTROLLERDEVICEREMAPPED: {
@@ -728,8 +741,7 @@ void WindowContext::ProcessEvent(double time_s) {
 	}
 }
 
-#if defined(__APPLE__)
-void WindowContext::RunOnMainThread(std::function<void()> task, bool wait) {
+void WindowContext::RunOnMainThread(std::function<void()> task) {
 	if (Common::Thread::IsMainThread()) {
 		task();
 		return;
@@ -747,9 +759,6 @@ void WindowContext::RunOnMainThread(std::function<void()> task, bool wait) {
 	event.type = SDL_USEREVENT;
 	SDL_PushEvent(&event);
 
-	if (!wait) {
-		return;
-	}
 	Common::LockGuard lock(main_task_mutex);
 	while (main_tasks_run < ticket) {
 		main_task_done.Wait(&main_task_mutex);
@@ -772,7 +781,6 @@ void WindowContext::DrainMainThreadTasks() {
 	main_tasks_run += tasks.size();
 	main_task_done.SignalAll();
 }
-#endif
 
 void WindowContext::Run() {
 	Common::Timer timer;
@@ -783,9 +791,7 @@ void WindowContext::Run() {
 	loop.paused.store(false, std::memory_order_release);
 
 	while (!loop.need_exit) {
-#if defined(__APPLE__)
 		DrainMainThreadTasks();
-#endif
 		if (loop.paused.load(std::memory_order_acquire)) {
 			if (!timer.IsPaused()) {
 				timer.Pause();
@@ -794,8 +800,8 @@ void WindowContext::Run() {
 			timer.Resume();
 		}
 
-		if (SDL_WaitEvent(&loop.event) == 0) {
-			EXIT("%s\n", SDL_GetError());
+		if (!HostInputWaitEvent(&loop.event)) {
+			continue;
 		}
 		ProcessEvent(timer.GetTimeS());
 	}
@@ -821,11 +827,26 @@ static void WindowCreate(WindowContext& context) {
 
 	LOGF("WindowCreate(): width = %d, height = %d\n", width, height);
 
-	uint32_t window_flags = KYTY_SDL_WINDOW_FLAGS;
-	if (Config::FullscreenEnabled()) {
-		window_flags |= static_cast<uint32_t>(SDL_WINDOW_FULLSCREEN_DESKTOP);
-	}
+	uint32_t window_flags = WindowContext::InitialWindowFlags(Config::FullscreenEnabled());
 #if defined(__APPLE__)
+	// SDL loads Vulkan while creating a Vulkan window, so select the bundled
+	// MoltenVK loader before calling SDL_CreateWindow. Keep an explicit user
+	// override, and fall back to SDL's normal loader search when no bundle is present.
+	if (std::getenv("SDL_VULKAN_LIBRARY") == nullptr) {
+		if (char* base_path = SDL_GetBasePath(); base_path != nullptr) {
+			const std::string base_path_str = base_path;
+			SDL_free(base_path);
+			std::string moltenvk_path = base_path_str + "libMoltenVK.dylib";
+			if (!Common::File::IsFileExisting(moltenvk_path)) {
+				moltenvk_path = base_path_str + "../Frameworks/libMoltenVK.dylib";
+			}
+			if (Common::File::IsFileExisting(moltenvk_path) &&
+			    SDL_setenv("SDL_VULKAN_LIBRARY", moltenvk_path.c_str(), 0) == 0) {
+				LOGF("Vulkan loader: %s\n", moltenvk_path.c_str());
+			}
+		}
+	}
+
 	// macOS 26 window chrome (CoreUI asset decode, SwiftUI titlebar) has been observed
 	// throwing NSExceptions under Rosetta during the first CATransaction commit. A
 	// borderless window skips that machinery entirely.
@@ -836,13 +857,20 @@ static void WindowCreate(WindowContext& context) {
 	context.window = SDL_CreateWindow(KYTY_SDL_WINDOW_CAPTION, KYTY_SDL_WINDOWPOS_CENTERED,
 	                                  KYTY_SDL_WINDOWPOS_CENTERED, width, height, window_flags);
 
-	context.window_hidden = true;
-
 	if (context.window == nullptr) {
 		EXIT("%s\n", SDL_GetError());
 	}
 
 	SDL_SetWindowResizable(context.window, SDL_FALSE);
+	context.UpdateIcon();
+}
+
+uint32_t WindowContext::InitialWindowFlags(bool fullscreen) noexcept {
+	auto flags = static_cast<uint32_t>(SDL_WINDOW_VULKAN);
+	if (fullscreen) {
+		flags |= static_cast<uint32_t>(SDL_WINDOW_FULLSCREEN_DESKTOP);
+	}
+	return flags;
 }
 
 Presenter& WindowInit(uint32_t width, uint32_t height) {
@@ -866,6 +894,7 @@ void WindowRun() {
 	EXIT_IF(g_window == nullptr);
 
 	g_window->Run();
+	g_window->render_context->GetPipelineCache().Save();
 }
 
 void WindowShutdown() {
@@ -959,6 +988,7 @@ void WindowContext::UpdateTitle() {
 	    Loader::SystemContentParamSfoGetString("TITLE_ID", title_id, sizeof(title_id));
 	static bool has_app_ver =
 	    Loader::SystemContentParamSfoGetString("APP_VER", app_ver, sizeof(app_ver));
+	static const std::string processor_name = Common::GetSystemInfo().ProcessorName;
 	static uint64_t fps_start   = Common::Timer::QueryPerformanceCounter();
 	static uint64_t frame_num   = 0;
 	static uint64_t fps_frames  = 0;
@@ -983,19 +1013,14 @@ void WindowContext::UpdateTitle() {
 		fps_frames  = 0;
 	}
 
-	auto fps = fmt::format(
+	const auto* device_name = graphic_ctx.GetPhysicalDeviceProperties().deviceName.data();
+	auto text = fmt::format(
 	    "[{} | {}] {}{}{}{}{}{}[{}] [{}], frame: {}, fps: {:f}", KYTY_BUILD_LABEL, build_type,
 	    (has_title ? title : ""), (has_title ? ", " : ""), (has_title_id ? title_id : ""),
 	    (has_title_id ? ", " : ""), (has_app_ver ? app_ver : ""), (has_app_ver ? " " : ""),
 	    device_name, processor_name, frame_num, current_fps);
 
-#if defined(__APPLE__)
-	// AppKit traps on title changes off the main thread; fire-and-forget keeps present pacing.
-	RunOnMainThread([this, fps = std::move(fps)] { SDL_SetWindowTitle(window, fps.c_str()); },
-	                false);
-#else
-	SDL_SetWindowTitle(window, fps.c_str());
-#endif
+	RunOnMainThread([this, text = std::move(text)] { SDL_SetWindowTitle(window, text.c_str()); });
 }
 
 } // namespace Libs::Graphics

src/graphics/presentation/window/windowInternal.h

@@ -9,13 +9,10 @@
 
 #include <atomic>
 #include <cstdint>
+#include <functional>
 #include <memory>
 #include <vector>
 
-#if defined(__APPLE__)
-#include <functional>
-#endif
-
 namespace Libs::Graphics {
 
 class Presenter;
@@ -39,6 +36,7 @@ struct WindowContext {
 	KYTY_CLASS_NO_COPY(WindowContext);
 
 	[[nodiscard]] static vk::PhysicalDeviceVulkan13Features RequiredVulkan13Features() noexcept;
+	[[nodiscard]] static uint32_t InitialWindowFlags(bool fullscreen) noexcept;
 	void                                                    CreateVulkan();
 	void                                                    RecreateSurface();
 	void                                                    RefreshSurfaceCapabilities();
@@ -49,36 +47,25 @@ struct WindowContext {
 	void ProcessDisplayEvent(const SDL_DisplayEvent& event);
 	void ProcessEvent(double time_seconds);
 	void Run();
-
-#if defined(__APPLE__)
-	// AppKit only allows window operations (show, icon, title, view/layer changes) on the
-	// main thread; the present thread marshals them through the SDL main loop with these.
-	// wait=true blocks until the task has run on the main thread.
-	void RunOnMainThread(std::function<void()> task, bool wait);
+	// SDL window operations must complete on the main thread.
+	void RunOnMainThread(std::function<void()> task);
 	void DrainMainThreadTasks();
-#endif
 
 	GraphicContext                 graphic_ctx;
 	SDL_Window*                    window        = nullptr;
-	bool                           window_hidden = true;
 	vk::SurfaceKHR                 surface       = nullptr;
 	SurfaceCapabilities            surface_capabilities;
 	std::unique_ptr<RenderContext> render_context;
 	std::unique_ptr<Presenter>     presenter;
 	WindowLoopState                loop;
 
-	char device_name[VK_MAX_PHYSICAL_DEVICE_NAME_SIZE] = {0};
-	char processor_name[64]                            = {0};
-
 	Common::Mutex mutex;
 
-#if defined(__APPLE__)
 	Common::Mutex                      main_task_mutex;
 	Common::CondVar                    main_task_done;
 	std::vector<std::function<void()>> main_tasks;            // guarded by main_task_mutex
 	uint64_t                           main_tasks_queued = 0; // guarded by main_task_mutex
 	uint64_t                           main_tasks_run    = 0; // guarded by main_task_mutex
-#endif
 };
 
 } // namespace Libs::Graphics

src/graphics/shader/recompiler/ShaderRecompiler.cpp

@@ -7,7 +7,6 @@
 #include "graphics/shader/recompiler/frontend/decode/ShaderDecoder.h"
 #include "graphics/shader/recompiler/frontend/translate/Translate.h"
 #include "graphics/shader/recompiler/ir/ShaderIR.h"
-#include "graphics/shader/recompiler/ir/ValueProgram.h"
 #include "graphics/shader/recompiler/ir/passes/BindingLayout.h"
 #include "graphics/shader/recompiler/ir/passes/ConstantPropagation.h"
 #include "graphics/shader/recompiler/ir/passes/DeadCodeElimination.h"
@@ -15,6 +14,7 @@
 #include "graphics/shader/recompiler/ir/passes/ResourceMaterialization.h"
 #include "graphics/shader/recompiler/ir/passes/ResourceTracking.h"
 #include "graphics/shader/recompiler/ir/passes/ShaderInfoCollection.h"
+#include "graphics/shader/recompiler/ir/passes/SharedMemoryBarrier.h"
 #include "graphics/shader/recompiler/ir/passes/SrtWalker.h"
 #include "graphics/shader/recompiler/ir/passes/SsaRewrite.h"
 
@@ -30,14 +30,6 @@ namespace Libs::Graphics::ShaderRecompiler {
 
 namespace {
 
-bool ReadZeroMemory(void*, uint64_t, uint32_t* value) {
-	if (value == nullptr) {
-		return false;
-	}
-	*value = 0;
-	return true;
-}
-
 const char* GetDumpLabel(const CompileOptions& options) {
 	return options.dump_label != nullptr ? options.dump_label : "ShaderRecompiler";
 }
@@ -48,7 +40,7 @@ std::string MakeIrDump(const CFG::Graph& cfg, const IR::Program& ir) {
 	dump += "\nIR:\n";
 	dump += fmt::format("mode={} scratch_dwords={}\n",
 	                    ir.dispatcher_fallback ? "dispatcher" : "structured", ir.scratch_dwords);
-	dump += ir.values != nullptr ? IR::ValueProgramToString(*ir.values) : IR::ProgramToString(ir);
+	dump += IR::ProgramToString(ir);
 	return dump;
 }
 
@@ -116,17 +108,8 @@ void ClearEmbeddedFetchVectorLanes(EmbeddedFetchVectorLanes* lanes, uint32_t reg
 	lanes->erase(first, last);
 }
 
-struct EmbeddedFetchLoad {
-	uint32_t              pc         = 0;
-	int                   attrib_id  = -1;
-	uint32_t              components = 0;
-	std::vector<uint32_t> prolog_loads;
-};
-
-struct EmbeddedFetchData {
-	std::vector<EmbeddedFetchLoad> loads;
-	int32_t                        vertex_offset_sgpr = -1;
-};
+using EmbeddedFetchLoad = Frontend::EmbeddedFetchLoad;
+using EmbeddedFetchData = Frontend::EmbeddedFetchPlan;
 
 bool IsDecodedSgpr(const Decoder::Operand& op) {
 	return op.kind == Decoder::OperandKind::Sgpr || op.kind == Decoder::OperandKind::VccLo ||
@@ -245,15 +228,6 @@ int BufferTableAttribFromOffset(uint32_t raw_offset, int dword) {
 	return static_cast<int>((raw_offset + static_cast<uint32_t>(dword) * 4u) / 16u);
 }
 
-void AppendUniquePcs(std::vector<uint32_t>& dst, const std::vector<uint32_t>& src) {
-	dst.reserve(dst.size() + src.size());
-	for (auto pc: src) {
-		if (std::find(dst.begin(), dst.end(), pc) == dst.end()) {
-			dst.push_back(pc);
-		}
-	}
-}
-
 EmbeddedFetchData DetectEmbeddedVertexFetch(const Decoder::Program&      decoded,
                                             const ShaderVertexInputInfo* input_info,
                                             uint32_t user_data_base, uint32_t user_data_count,
@@ -488,119 +462,16 @@ EmbeddedFetchData DetectEmbeddedVertexFetch(const Decoder::Program&      decoded
 	return data;
 }
 
-const EmbeddedFetchLoad* FindEmbeddedFetchLoad(const std::vector<EmbeddedFetchLoad>& loads,
-                                               uint32_t                              pc) {
-	for (const auto& load: loads) {
-		if (load.pc == pc) {
-			return &load;
-		}
-	}
-	return nullptr;
-}
-
-bool EmbeddedFetchPcInList(const std::vector<uint32_t>& pcs, uint32_t pc) {
-	return std::find(pcs.begin(), pcs.end(), pc) != pcs.end();
-}
-
-bool IsIrFetchPrologLoad(const IR::Instruction& inst) {
-	return inst.op == IR::Opcode::SLoadDword || inst.op == IR::Opcode::SBufferLoadDword;
-}
-
-int ResolveEmbeddedFetchResource(const ShaderVertexInputInfo* input_info,
-                                 const EmbeddedFetchLoad&     load) {
-	if (load.attrib_id >= 0 && load.attrib_id < input_info->resources_num &&
-	    input_info->resources_dst[load.attrib_id].attr_id == load.attrib_id) {
-		return load.attrib_id;
-	}
-	for (int i = 0; i < input_info->resources_num; i++) {
-		const auto& dst = input_info->resources_dst[i];
-		if (dst.attr_id == load.attrib_id &&
-		    load.components <= static_cast<uint32_t>(std::max(dst.registers_num, 1))) {
-			return i;
-		}
-	}
-	for (int i = 0; i < input_info->resources_num; i++) {
-		if (input_info->resources_dst[i].attr_id == load.attrib_id) {
-			return i;
-		}
-	}
-	return -1;
-}
-
-uint32_t RewriteEmbeddedVertexFetches(IR::Program& ir, const ShaderVertexInputInfo* input_info,
-                                      const std::vector<EmbeddedFetchLoad>& loads) {
-	if (loads.empty()) {
-		return 0;
-	}
-
-	std::vector<uint32_t> prolog_pcs;
-	prolog_pcs.reserve(loads.size());
-	for (const auto& load: loads) {
-		AppendUniquePcs(prolog_pcs, load.prolog_loads);
-	}
-
-	auto*    mutable_input_info = const_cast<ShaderVertexInputInfo*>(input_info);
-	uint32_t rewritten          = 0;
-	for (auto& block: ir.blocks) {
-		for (auto& inst: block.instructions) {
-			if (IsIrFetchPrologLoad(inst) && EmbeddedFetchPcInList(prolog_pcs, inst.pc)) {
-				auto pc = inst.pc;
-				inst    = {};
-				inst.pc = pc;
-				inst.op = IR::Opcode::ControlNop;
-				continue;
-			}
-
-			if (inst.op != IR::Opcode::BufferLoadDword) {
-				continue;
-			}
-			const auto* load = FindEmbeddedFetchLoad(loads, inst.pc);
-			if (load == nullptr || inst.memory.data_dwords != load->components ||
-			    inst.dst.kind != IR::OperandKind::Register ||
-			    inst.dst.reg.file != IR::RegisterFile::Vector) {
-				continue;
-			}
-
-			const auto resource_id = ResolveEmbeddedFetchResource(input_info, *load);
-			if (resource_id < 0 || resource_id >= input_info->resources_num) {
-				LOGF("ShaderRecompiler VS embedded fetch remap failed: pc=0x%08" PRIx32
-				     " attrib=%d resources=%d\n",
-				     inst.pc, load->attrib_id, input_info->resources_num);
-				continue;
-			}
-
-			inst.op                         = IR::Opcode::LoadInputF32;
-			inst.input_info.attr            = static_cast<uint32_t>(resource_id);
-			inst.input_info.chan            = 0;
-			inst.input_info.component_count = load->components;
-			inst.memory                     = {};
-			inst.src_count                  = 0;
-			mutable_input_info->resource_fetch_components[resource_id] =
-			    std::max(mutable_input_info->resource_fetch_components[resource_id],
-			             static_cast<int>(load->components));
-			rewritten += load->components;
-		}
-	}
-
-	return rewritten;
-}
-
 } // namespace
 
-bool TryRecompile(std::span<const uint32_t> code, const CompileOptions& options,
-                  CompileResult& result, std::string* error) {
+CompileResult Recompile(std::span<const uint32_t> code, const CompileOptions& options) {
 	if (code.empty()) {
-		if (error != nullptr) {
-			*error = "invalid shader recompiler input";
-		}
-		return false;
+		EXIT("shader recompiler input is empty\n");
 	}
 	if (options.stage != ShaderType::Compute && options.stage != ShaderType::Vertex &&
 	    options.stage != ShaderType::Pixel) {
-		if (error != nullptr) {
-			*error = "shader recompiler supports compute, vertex, and pixel stages";
-		}
-		return false;
+		EXIT("shader recompiler received unsupported stage %u\n",
+		     static_cast<unsigned>(options.stage));
 	}
 
 	const auto compile_begin = std::chrono::steady_clock::now();
@@ -615,9 +486,7 @@ bool TryRecompile(std::span<const uint32_t> code, const CompileOptions& options,
 	     static_cast<uint64_t>(code.size()));
 
 	Decoder::Program decoded;
-	if (!Decoder::DecodeProgram(code, decoded, error)) {
-		return false;
-	}
+	Decoder::DecodeProgram(code, decoded);
 	LOGF("%s phase end: stage=%s hash=0x%016" PRIx64 " decode instructions=%" PRIu64
 	     " elapsed_ms=%" PRIu64 "\n",
 	     GetDumpLabel(options), StageName(options.stage), options.shader_hash,
@@ -631,12 +500,9 @@ bool TryRecompile(std::span<const uint32_t> code, const CompileOptions& options,
 		}
 	}
 
-	CFG::Graph cfg;
 	LOGF("%s phase begin: stage=%s hash=0x%016" PRIx64 " CFG BuildGraph\n", GetDumpLabel(options),
 	     StageName(options.stage), options.shader_hash);
-	if (!CFG::BuildGraph(decoded, cfg, error)) {
-		return false;
-	}
+	auto cfg = CFG::BuildGraph(decoded);
 	LOGF("%s phase end: stage=%s hash=0x%016" PRIx64 " CFG BuildGraph blocks=%" PRIu64
 	     " loops=%" PRIu64 " back_edges=%" PRIu64 " elapsed_ms=%" PRIu64 "\n",
 	     GetDumpLabel(options), StageName(options.stage), options.shader_hash,
@@ -649,13 +515,12 @@ bool TryRecompile(std::span<const uint32_t> code, const CompileOptions& options,
 		dispatcher_reason   = cfg.unsupported_reason;
 		LogDispatcherFallback(options, cfg, "build", dispatcher_reason);
 	} else {
-		std::string structure_error;
-		const auto  unstructured_cfg = cfg;
+		const auto unstructured_cfg = cfg;
 		LOGF("%s phase begin: stage=%s hash=0x%016" PRIx64 " CFG Structurize\n",
 		     GetDumpLabel(options), StageName(options.stage), options.shader_hash);
-		if (!CFG::Structurize(cfg, &structure_error)) {
+		if (!CFG::Structurize(cfg)) {
 			dispatcher_fallback      = true;
-			dispatcher_reason        = structure_error;
+			dispatcher_reason        = cfg.unsupported_reason;
 			const auto failure_kind  = cfg.failure_kind;
 			const auto failure_block = cfg.failure_block;
 			LogDispatcherFallback(options, cfg, "structurize", dispatcher_reason);
@@ -663,7 +528,7 @@ bool TryRecompile(std::span<const uint32_t> code, const CompileOptions& options,
 			cfg.unsupported        = true;
 			cfg.failure_kind       = failure_kind;
 			cfg.failure_block      = failure_block;
-			cfg.unsupported_reason = structure_error;
+			cfg.unsupported_reason = dispatcher_reason;
 		} else {
 			LOGF("%s structured CFG success: blocks=%" PRIu64 "\n", GetDumpLabel(options),
 			     static_cast<uint64_t>(cfg.blocks.size()));
@@ -675,139 +540,105 @@ bool TryRecompile(std::span<const uint32_t> code, const CompileOptions& options,
 		     static_cast<uint64_t>(cfg.natural_loops.size()), phase_ms());
 	}
 
-	IR::Program ir;
-	LOGF("%s phase begin: stage=%s hash=0x%016" PRIx64 " IR LowerProgram\n", GetDumpLabel(options),
-	     StageName(options.stage), options.shader_hash);
-	const auto lower_program = [&](IR::Program& output) {
-		output = {};
-		if (!IR::LowerProgram(decoded, cfg, options.stage, options.wave_size, output, error)) {
-			return false;
-		}
-		output.lane_mask_mode  = options.lane_mask_mode;
-		output.shader_hash     = options.shader_hash;
-		output.user_data_base  = options.user_data_base;
-		output.user_data_count = options.user_data_count;
-		output.scratch_dwords  = options.scratch_dwords;
-		return true;
-	};
-	if (!lower_program(ir)) {
-		return false;
-	}
-
-	LOGF("%s phase end: stage=%s hash=0x%016" PRIx64 " IR LowerProgram blocks=%" PRIu64
-	     " elapsed_ms=%" PRIu64 "\n",
-	     GetDumpLabel(options), StageName(options.stage), options.shader_hash,
-	     static_cast<uint64_t>(ir.blocks.size()), phase_ms());
 	const ShaderVertexInputInfo*  vertex  = nullptr;
 	const ShaderPixelInputInfo*   pixel   = nullptr;
 	const ShaderComputeInputInfo* compute = nullptr;
 	switch (options.stage) {
-		case ShaderType::Vertex: vertex = options.input_info.vertex; break;
-		case ShaderType::Pixel: pixel = options.input_info.pixel; break;
-		case ShaderType::Compute: compute = options.input_info.compute; break;
+		case ShaderType::Vertex:
+			vertex = options.input_info.vertex;
+			break;
+		case ShaderType::Pixel:
+			pixel = options.input_info.pixel;
+			break;
+		case ShaderType::Compute:
+			compute = options.input_info.compute;
+			break;
 		default: break;
 	}
 	EmbeddedFetchData embedded_fetch;
 	if (options.stage == ShaderType::Vertex && vertex->fetch_embedded) {
-		embedded_fetch = DetectEmbeddedVertexFetch(decoded, vertex, ir.user_data_base,
-		                                           ir.user_data_count, options.wave_size);
-		auto rewritten = RewriteEmbeddedVertexFetches(ir, vertex, embedded_fetch.loads);
-		if (rewritten > 0 || !embedded_fetch.loads.empty()) {
-			LOGF("%s embedded vertex fetch rewrite: detected=%" PRIu64 " rewritten=%" PRIu32 "\n",
-			     GetDumpLabel(options), static_cast<uint64_t>(embedded_fetch.loads.size()),
-			     rewritten);
+		embedded_fetch = DetectEmbeddedVertexFetch(decoded, vertex, options.user_data_base,
+		                                           static_cast<uint32_t>(options.user_data.size()),
+		                                           options.wave_size);
+		if (!embedded_fetch.loads.empty()) {
+			LOGF("%s embedded vertex fetch plan: detected=%" PRIu64 "\n", GetDumpLabel(options),
+			     static_cast<uint64_t>(embedded_fetch.loads.size()));
 		}
 	}
-	ir.dispatcher_fallback = dispatcher_fallback;
-	if (!dispatcher_reason.empty()) {
-		ir.fallback_reason = dispatcher_reason;
-	}
-
-	ir.values = std::make_shared<IR::ValueProgram>();
-	if (!Frontend::TranslateProgram(ir, *ir.values, vertex, pixel, compute, error)) {
-		return false;
-	}
-	ir.blocks.clear();
-	IR::RewriteToSsa(ir.values->blocks);
-	IR::ConstantPropagationPass(ir.values->blocks);
-	for (auto& info: ir.values->block_info) {
-		info.condition       = info.condition.Resolve();
-		info.indirect_target = info.indirect_target.Resolve();
-	}
-	IR::RemoveIdentities(ir.values->blocks);
-	IR::EliminateDeadCode(ir.values->blocks);
-	const auto read_lane_stats = IR::EliminateReadLane(*ir.values, ir.wave_size);
+	Frontend::TranslateOptions translate_options {
+	    .stage               = options.stage,
+	    .wave_size           = options.wave_size,
+	    .shader_hash         = options.shader_hash,
+	    .user_data_base      = options.user_data_base,
+	    .user_data_count     = static_cast<uint32_t>(options.user_data.size()),
+	    .scratch_dwords      = options.scratch_dwords,
+	    .dispatcher_fallback = dispatcher_fallback,
+	    .cfg_failure_kind    = cfg.failure_kind,
+	    .fallback_reason     = dispatcher_reason.empty() ? cfg.unsupported_reason
+	                                                    : dispatcher_reason,
+	    .vertex              = vertex,
+	    .pixel               = pixel,
+	    .compute             = compute,
+	    .embedded_fetch      = embedded_fetch.loads.empty() ? nullptr : &embedded_fetch,
+	};
+	LOGF("%s phase begin: stage=%s hash=0x%016" PRIx64 " IR TranslateProgram\n",
+	     GetDumpLabel(options), StageName(options.stage), options.shader_hash);
+	auto ir = Frontend::TranslateProgram(decoded, cfg, translate_options);
+	LOGF("%s phase end: stage=%s hash=0x%016" PRIx64 " IR TranslateProgram blocks=%" PRIu64
+	     " elapsed_ms=%" PRIu64 "\n",
+	     GetDumpLabel(options), StageName(options.stage), options.shader_hash,
+	     static_cast<uint64_t>(ir.blocks.size()), phase_ms());
+	IR::RewriteToSsa(ir.blocks);
+	IR::ConstantPropagationPass(ir.blocks);
+	IR::ResolveControlFlowIdentities(ir);
+	IR::RemoveIdentities(ir.blocks);
+	IR::EliminateDeadCode(ir.blocks);
+	const auto read_lane_stats = IR::EliminateReadLane(ir, ir.wave_size);
 	if (read_lane_stats.rewritten_reads != 0) {
 		LOGF("%s read-lane elimination: reads=%" PRIu32 "\n", GetDumpLabel(options),
 		     read_lane_stats.rewritten_reads);
-		IR::ConstantPropagationPass(ir.values->blocks);
-		IR::RemoveIdentities(ir.values->blocks);
-		IR::EliminateDeadCode(ir.values->blocks);
+		IR::ConstantPropagationPass(ir.blocks);
+		IR::ResolveControlFlowIdentities(ir);
+		IR::RemoveIdentities(ir.blocks);
+		IR::EliminateDeadCode(ir.blocks);
 	}
-	std::string srt_error;
-	if (!IR::BuildSrtPlan(ir, &srt_error)) {
-		LOGF("%s SRT planning failed: %s\n", GetDumpLabel(options), srt_error.c_str());
-		if (error != nullptr) {
-			*error = std::move(srt_error);
+	if (options.stage == ShaderType::Compute) {
+		const auto lds_barriers =
+		    IR::InsertSharedMemoryBarriers(ir, ir.wave_size, *compute);
+		if (lds_barriers.inserted_barriers != 0) {
+			LOGF("%s wave64 LDS synchronization: barriers=%" PRIu32 "\n", GetDumpLabel(options),
+			     lds_barriers.inserted_barriers);
 		}
-		return false;
 	}
-	IR::EliminateDeadCode(ir.values->blocks);
-	if (!IR::TrackResources(ir, error)) {
-		return false;
-	}
-	IR::EliminateDeadCode(ir.values->blocks);
+	IR::BuildSrtPlan(ir);
+	IR::EliminateDeadCode(ir.blocks);
+	IR::TrackResources(ir);
+	IR::EliminateDeadCode(ir.blocks);
 	if (options.stage == ShaderType::Vertex) {
 		ir.info.vertex_offset_sgpr = embedded_fetch.vertex_offset_sgpr;
 	}
 
+	const IR::SrtRuntime runtime {
+	    .user_data                  = options.user_data,
+	    .shader_base                = reinterpret_cast<uint64_t>(code.data()),
+	    .read_memory                = options.read_memory,
+	    .userdata                   = options.read_memory_data,
+	    .read_specialization_memory = options.read_specialization_memory,
+	};
 	IR::ResourceSnapshot resources;
-	if (options.resource_snapshot != nullptr) {
-		resources = *options.resource_snapshot;
-	} else {
-		IR::SrtRuntime           runtime;
-		std::array<uint32_t, 64> zero_user_data {};
-		if (options.user_data == nullptr) {
-			for (uint32_t i = 2; i < zero_user_data.size(); i += 4) {
-				zero_user_data[i] = UINT32_MAX;
-			}
-		}
-		const auto* user_data =
-		    options.user_data != nullptr ? options.user_data : zero_user_data.data();
-		runtime.user_data   = std::span<const uint32_t>(user_data, options.user_data_count);
-		runtime.shader_base = options.shader_base != 0 ? options.shader_base
-		                                               : reinterpret_cast<uint64_t>(code.data());
-		runtime.read_memory = options.read_memory;
-		runtime.read_specialization_memory = options.read_specialization_memory;
-		if (runtime.read_memory == nullptr && options.user_data == nullptr) {
-			runtime.read_memory = ReadZeroMemory;
-		}
-		runtime.userdata         = options.read_memory_data;
-		runtime.flat_memory_base = options.flat_memory_base;
-		if (!IR::MaterializeResources(ir, runtime, resources, error)) {
-			return false;
-		}
-	}
-	if (!IR::SpecializeResources(ir, resources, error)) {
-		return false;
+	if (!IR::MaterializeResources(ir, runtime, resources)) {
+		EXIT("shader resource materialization failed: stage=%s hash=0x%016" PRIx64 "\n",
+		     StageName(options.stage), options.shader_hash);
 	}
+	IR::SpecializeResources(ir, resources);
 
 	IR::ShaderInfoOptions info_options;
 	info_options.vertex  = vertex;
 	info_options.pixel   = pixel;
 	info_options.compute = compute;
-	if (!IR::CollectShaderInfo(ir, info_options, error)) {
-		return false;
-	}
-	IR::BindingLayoutOptions layout_options;
-	layout_options.descriptor_set       = options.descriptor_set;
-	layout_options.push_constant_offset = options.push_constant_offset;
-	if (!IR::AllocateBindings(ir, layout_options, error)) {
-		return false;
-	}
-	if (!Spirv::AnalyzeProgramRequirements(ir, error)) {
-		return false;
-	}
+	IR::CollectShaderInfo(ir, info_options);
+	IR::AllocateBindings(ir, options.push_constant_offset);
+	Spirv::AnalyzeProgramRequirements(ir);
 	std::string ir_dump;
 	if (options.dump_ir) {
 		ir_dump = MakeIrDump(cfg, ir);
@@ -816,36 +647,22 @@ bool TryRecompile(std::span<const uint32_t> code, const CompileOptions& options,
 		}
 	}
 
-	std::vector<uint32_t> spirv;
-	std::string           emit_error;
 	LOGF("%s phase begin: stage=%s hash=0x%016" PRIx64 " SPIR-V EmitProgram\n",
 	     GetDumpLabel(options), StageName(options.stage), options.shader_hash);
-	if (!Spirv::EmitProgram(ir, resources, options.input_info, spirv, &emit_error)) {
-		LOGF("%s typed SPIR-V emit failed: %s\n", GetDumpLabel(options), emit_error.c_str());
-		if (dispatcher_fallback && error != nullptr) {
-			*error = fmt::format("dispatcher fallback failed after {}: {}",
-			                     dispatcher_reason.c_str(), emit_error.c_str());
-			LOGF("%s dispatcher fallback emit failed: %s\n", GetDumpLabel(options), error->c_str());
-		} else if (error != nullptr) {
-			*error = emit_error;
-		}
-		return false;
-	}
+	auto spirv = Spirv::EmitProgram(ir, resources, options.input_info);
 	LOGF("%s phase end: stage=%s hash=0x%016" PRIx64 " SPIR-V EmitProgram words=%" PRIu64
 	     " elapsed_ms=%" PRIu64 "\n",
 	     GetDumpLabel(options), StageName(options.stage), options.shader_hash,
 	     static_cast<uint64_t>(spirv.size()), phase_ms());
+	CompileResult result;
 	result.spirv     = std::move(spirv);
 	result.program   = std::move(ir);
 	result.resources = std::move(resources);
 	if (options.dump_ir) {
 		result.decoded_dump = std::move(decoded_dump);
 		result.ir_dump      = std::move(ir_dump);
-	} else {
-		result.decoded_dump.clear();
-		result.ir_dump.clear();
 	}
-	return true;
+	return result;
 }
 
 } // namespace Libs::Graphics::ShaderRecompiler

src/graphics/shader/recompiler/ShaderRecompiler.h

@@ -6,7 +6,6 @@
 #include "graphics/shader/recompiler/ir/passes/ResourceMaterialization.h"
 #include "graphics/shader/shader.h"
 
-#include <optional>
 #include <span>
 #include <vector>
 
@@ -14,24 +13,18 @@ namespace Libs::Graphics::ShaderRecompiler {
 
 struct CompileOptions {
 	ShaderType                  stage           = ShaderType::Compute;
-	ShaderLaneMaskMode          lane_mask_mode  = ShaderLaneMaskMode::NativeWave;
 	uint32_t                    wave_size       = 64;
 	uint32_t                    user_data_base  = 0;
-	uint32_t                    user_data_count = 64;
 	uint32_t                    scratch_dwords  = 0;
 	uint64_t                    shader_hash     = 0;
-	uint64_t                    shader_base     = 0;
-	std::optional<uint64_t>     flat_memory_base;
-	uint32_t                    descriptor_set             = 0;
-	uint32_t                    push_constant_offset       = 0;
+	uint32_t                    push_constant_offset      = 0;
 	bool                        dump_ir                    = true;
 	bool                        early_dump                 = false;
 	const char*                 dump_label                 = nullptr;
-	const uint32_t*             user_data                  = nullptr;
+	std::span<const uint32_t>   user_data;
 	IR::SrtMemoryReader         read_memory                = nullptr;
 	IR::SrtMemoryReader         read_specialization_memory = nullptr;
 	void*                       read_memory_data           = nullptr;
-	const IR::ResourceSnapshot* resource_snapshot          = nullptr;
 	ShaderStageInputInfo        input_info;
 };
 
@@ -43,8 +36,8 @@ struct CompileResult {
 	IR::ResourceSnapshot  resources;
 };
 
-bool TryRecompile(std::span<const uint32_t> code, const CompileOptions& options,
-                  CompileResult& result, std::string* error);
+[[nodiscard]] CompileResult Recompile(std::span<const uint32_t> code,
+                                      const CompileOptions& options);
 
 } // namespace Libs::Graphics::ShaderRecompiler
 

src/graphics/shader/recompiler/backend/spirv/SpirvEmitter.cpp

@@ -1,19 +1,23 @@
 #include "graphics/shader/recompiler/backend/spirv/SpirvEmitter.h"
 
+#include "common/assert.h"
 #include "graphics/shader/recompiler/backend/spirv/spirvEmitterInternal.h"
-#include "graphics/shader/recompiler/ir/ValueProgram.h"
-#include "graphics/shader/shader.h"
+#include "graphics/shader/recompiler/ir/ShaderIR.h"
 
 #include <algorithm>
 #include <array>
 
 namespace Libs::Graphics::ShaderRecompiler::Spirv {
 
-using ShaderError::Fail;
-
 namespace {
 
-bool ValidateNativeProgram(const IR::Program& program, std::string* error) {
+[[noreturn]] void Fail(const IR::Program& program, const char* reason) {
+	EXIT("SPIR-V validation failed: hash=0x%016" PRIx64 " stage=%u reason=%s\n",
+	     program.shader_hash, static_cast<unsigned>(program.stage), reason);
+	std::abort();
+}
+
+void ValidateNativeProgram(const IR::Program& program) {
 	using Kind                                             = IR::DescriptorBindingKind;
 	constexpr auto                               KindCount = static_cast<size_t>(Kind::Count);
 	std::array<std::vector<uint32_t>, KindCount> expected;
@@ -36,13 +40,13 @@ bool ValidateNativeProgram(const IR::Program& program, std::string* error) {
 	for (uint32_t i = 0; i < program.info.images.size(); i++) {
 		const auto kind = IR::DescriptorBindingForImage(program.info.images[i]);
 		if (!kind.has_value()) {
-			return Fail(error, "native shader plan has an invalid image class");
+			Fail(program, "native shader plan has an invalid image class");
 		}
 		present[static_cast<size_t>(*kind)] = true;
 		const auto dynamic = program.info.images[i].mip_mode == IR::ImageMipMode::DynamicStorage;
 		const auto count   = dynamic ? program.info.images[i].mip_count : 1u;
 		if (count == 0u || (!dynamic && program.info.images[i].mip_count != 1u)) {
-			return Fail(error, "native shader plan has an invalid image mip descriptor count");
+			Fail(program, "native shader plan has an invalid image mip descriptor count");
 		}
 		expected[static_cast<size_t>(*kind)].insert(expected[static_cast<size_t>(*kind)].end(),
 		                                            count, i);
@@ -51,36 +55,34 @@ bool ValidateNativeProgram(const IR::Program& program, std::string* error) {
 		Expect(Kind::Samplers, Dense(program.info.samplers.size()));
 	}
 	bool uses_gds = false;
-	if (program.values != nullptr) {
-		for (const auto* block: program.values->blocks) {
-			for (const auto& inst: *block) {
-				if (IR::SharedAccessOf(inst.GetOpcode()) == IR::SharedAccess::None) {
-					continue;
-				}
-				const auto index = inst.Flags<IR::MemoryFlags>().index;
-				if (index >= program.values->memory_info.size()) {
-					return Fail(error, "shared operation has invalid memory metadata");
-				}
-				const auto kind = program.values->memory_info[index].kind;
-				if (kind != IR::ResourceKind::Lds && kind != IR::ResourceKind::Gds) {
-					return Fail(error, "shared operation has invalid resource kind");
-				}
-				uses_gds |= kind == IR::ResourceKind::Gds;
+	for (const auto* block: program.blocks) {
+		for (const auto& inst: *block) {
+			if (IR::SharedAccessOf(inst.GetOpcode()) == IR::SharedAccess::None) {
+				continue;
 			}
+			const auto index = inst.Flags<IR::MemoryFlags>().index;
+			if (index >= program.memory_info.size()) {
+				Fail(program, "shared operation has invalid memory metadata");
+			}
+			const auto kind = program.memory_info[index].kind;
+			if (kind != IR::ResourceKind::Lds && kind != IR::ResourceKind::Gds) {
+				Fail(program, "shared operation has invalid resource kind");
+			}
+			uses_gds |= kind == IR::ResourceKind::Gds;
 		}
 	}
 	if (uses_gds) {
 		Expect(Kind::Gds);
 	}
-	if (!program.info.addresses.empty()) {
-		Expect(Kind::AddressMemory, Dense(program.info.addresses.size()));
+	if (program.info.uses_dma) {
+		Expect(Kind::BdaPagetable);
+		Expect(Kind::FaultBuffer);
 	}
 	const bool uses_flattened_runtime =
-	    program.values != nullptr &&
-	    (!program.values->srt_reads.empty() ||
+	    !program.srt_reads.empty() ||
 	     std::ranges::any_of(program.info.images, [](const IR::ImageResource& image) {
 		     return image.indirect_mapping_capacity != 0u;
-	     }));
+	     });
 	if (uses_flattened_runtime) {
 		Expect(Kind::FlattenedSrt);
 	}
@@ -89,29 +91,25 @@ bool ValidateNativeProgram(const IR::Program& program, std::string* error) {
 	}
 
 	std::array<bool, KindCount> seen {};
-	for (uint32_t i = 0; i < program.bindings.descriptors.size(); i++) {
-		const auto& binding = program.bindings.descriptors[i];
-		const auto  kind    = static_cast<size_t>(binding.kind);
+	for (const auto& binding: program.bindings.descriptors) {
+		const auto kind = static_cast<size_t>(binding.kind);
 		if (kind >= KindCount || seen[kind] || !present[kind] ||
 		    binding.resources != expected[kind]) {
-			return Fail(error, "native descriptor groups do not match shader topology");
-		}
-		for (uint32_t j = 0; j < i; j++) {
-			if (program.bindings.descriptors[j].binding == binding.binding) {
-				return Fail(error, "native descriptor binding numbers are not unique");
-			}
+			Fail(program, "native descriptor groups do not match shader topology");
 		}
 		seen[kind] = true;
 	}
 	for (size_t i = 0; i < KindCount; i++) {
 		if (present[i] != seen[i]) {
-			return Fail(error, "native shader plan is missing a required descriptor group");
+			Fail(program, "native shader plan is missing a required descriptor group");
 		}
 	}
 	const auto has_user_storage = present[static_cast<size_t>(Kind::UserData)];
-	if (program.bindings.memory_offset_dword != program.bindings.user_data_registers.size() ||
-	    program.bindings.memory_offset_count !=
-	        program.info.buffers.size() + program.info.addresses.size() ||
+	if (program.bindings.push_constant_offset % sizeof(uint32_t) != 0 ||
+	    program.bindings.push_constant_offset + program.bindings.push_constant_size >
+	        IR::NativePushConstantSize ||
+	    program.bindings.memory_offset_dword != program.bindings.user_data_registers.size() ||
+	    program.bindings.memory_offset_count != program.info.buffers.size() ||
 	    (!has_user_storage &&
 	     program.bindings.push_constant_size != program.bindings.ShaderDataDwords() * 4u) ||
 	    (has_user_storage && program.bindings.push_constant_size != 0) ||
@@ -120,12 +118,9 @@ bool ValidateNativeProgram(const IR::Program& program, std::string* error) {
 	    std::adjacent_find(program.bindings.user_data_registers.begin(),
 	                       program.bindings.user_data_registers.end()) !=
 	        program.bindings.user_data_registers.end()) {
-		return Fail(error, "native user-data layout is inconsistent");
+		Fail(program, "native user-data layout is inconsistent");
 	}
 
-	if (program.values == nullptr) {
-		return Fail(error, "native shader plan has no typed SSA");
-	}
 	const auto planning_only_handle = [&](const IR::Inst& handle) {
 		return !handle.Uses().empty() &&
 		       std::ranges::all_of(handle.Uses(), [&](const IR::Use& use) {
@@ -135,11 +130,11 @@ bool ValidateNativeProgram(const IR::Program& program, std::string* error) {
 				       return false;
 			       }
 			       const auto index = use.user->Flags<IR::MemoryFlags>().index;
-			       return index < program.values->memory_info.size() &&
-			              program.values->memory_info[index].planning_only;
+			       return index < program.memory_info.size() &&
+			              program.memory_info[index].planning_only;
 		       });
 	};
-	for (const auto* block: program.values->blocks) {
+	for (const auto* block: program.blocks) {
 		for (const auto& inst: *block) {
 			const auto dense = inst.Flags<uint32_t>();
 			switch (inst.GetOpcode()) {
@@ -148,37 +143,37 @@ bool ValidateNativeProgram(const IR::Program& program, std::string* error) {
 						break;
 					}
 					if (dense >= program.info.buffers.size()) {
-						return Fail(error, "typed buffer handle has an invalid dense resource");
+						Fail(program, "typed buffer handle has an invalid dense resource");
 					}
 					break;
 				case IR::ValueOpcode::GetAddressResource:
 					if (planning_only_handle(inst)) {
 						break;
 					}
-					if (dense >= program.info.addresses.size()) {
-						return Fail(error, "typed address handle has an invalid dense resource");
+					if (inst.NumArgs() != 2 || !program.info.uses_dma) {
+						Fail(program, "typed address handle has invalid DMA metadata");
 					}
 					break;
 				case IR::ValueOpcode::GetScratchResource:
 					if (inst.NumArgs() != 0 || program.scratch_dwords == 0) {
-						return Fail(error, "typed scratch handle has invalid shader metadata");
+						Fail(program, "typed scratch handle has invalid shader metadata");
 					}
 					break;
 				case IR::ValueOpcode::GetImageResource:
 					if (dense >= program.info.images.size()) {
-						return Fail(error, "typed image handle has an invalid dense resource");
+						Fail(program, "typed image handle has an invalid dense resource");
 					}
 					break;
 				case IR::ValueOpcode::GetSamplerResource:
 					if (dense >= program.info.samplers.size()) {
-						return Fail(error, "typed sampler handle has an invalid dense resource");
+						Fail(program, "typed sampler handle has an invalid dense resource");
 					}
 					break;
 				case IR::ValueOpcode::ReadConst: {
 					const auto slot = inst.Arg(1).Resolve();
 					if (!slot.IsImmediate() || slot.GetType() != IR::Type::U32 ||
-					    slot.U32() >= program.values->srt_reads.size()) {
-						return Fail(error, "flattened SRT read has an invalid dense slot");
+					    slot.U32() >= program.srt_reads.size()) {
+						Fail(program, "flattened SRT read has an invalid dense slot");
 					}
 					break;
 				}
@@ -186,51 +181,45 @@ bool ValidateNativeProgram(const IR::Program& program, std::string* error) {
 			}
 		}
 	}
-	return true;
 }
 
 } // namespace
 
-bool AnalyzeProgramRequirements(IR::Program& program, std::string* error) {
+void AnalyzeProgramRequirements(IR::Program& program) {
 	program.spirv_requirements.reset();
 	IR::SpirvRequirements requirements {};
-	if (program.values == nullptr) {
-		program.spirv_requirements.emplace(requirements);
-		return true;
-	}
-	const auto MarkExactSubgroup = [&] {
-		requirements.requires_exact_subgroup = true;
-		requirements.subgroup_ballot         = true;
-	};
-	for (const auto* block: program.values->blocks) {
+	const auto MarkBallot = [&] { requirements.subgroup_ballot = true; };
+	for (const auto* block: program.blocks) {
 		for (const auto& inst: *block) {
-			if (IR::AddressOpcodeInfoOf(inst.GetOpcode()).access != IR::AddressAccess::None) {
+			const auto address_access = IR::AddressOpcodeInfoOf(inst.GetOpcode()).access;
+			if (address_access != IR::AddressAccess::None) {
 				const auto memory_index = inst.Flags<IR::MemoryFlags>().index;
-				if (memory_index >= program.values->memory_info.size()) {
-					return Fail(error, "address operation has invalid memory metadata");
+				if (memory_index >= program.memory_info.size()) {
+					Fail(program, "address operation has invalid memory metadata");
 				}
-				if (program.values->memory_info[memory_index].kind == IR::ResourceKind::Scratch) {
+				if (program.memory_info[memory_index].kind == IR::ResourceKind::Scratch) {
 					if (program.scratch_dwords == 0) {
-						return Fail(error, "scratch operation has no per-thread storage");
+						Fail(program, "scratch operation has no per-thread storage");
 					}
 					requirements.function_scratch = true;
+				} else if (address_access == IR::AddressAccess::Write) {
+					Fail(program, "writable FLAT/GLOBAL addresses require GPU ownership tracking");
 				}
 			}
 			if (IR::BufferAccessOf(inst.GetOpcode()) != IR::BufferAccess::None) {
 				const auto memory_index = inst.Flags<IR::MemoryFlags>().index;
-				if (memory_index >= program.values->memory_info.size()) {
-					return Fail(error, "buffer operation has invalid memory metadata");
+				if (memory_index >= program.memory_info.size()) {
+					Fail(program, "buffer operation has invalid memory metadata");
 				}
-				const auto& memory = program.values->memory_info[memory_index];
+				const auto& memory = program.memory_info[memory_index];
 				if (memory.kind == IR::ResourceKind::Buffer) {
 					if (memory.resource >= program.info.buffers.size()) {
-						return Fail(error, "buffer operation has invalid resource metadata");
+						Fail(program, "buffer operation has invalid resource metadata");
 					}
 					if ((program.info.buffers[memory.resource].packed_stride & (1u << 20u)) != 0u) {
 						if (program.stage != ShaderType::Compute) {
-							return Fail(error, "buffer ADD_TID is only valid for compute shaders");
+							Fail(program, "buffer ADD_TID is only valid for compute shaders");
 						}
-						requirements.requires_exact_subgroup      = true;
 						requirements.subgroup_local_invocation_id = true;
 					}
 				}
@@ -238,29 +227,29 @@ bool AnalyzeProgramRequirements(IR::Program& program, std::string* error) {
 			const auto shared_access = IR::SharedAccessOf(inst.GetOpcode());
 			if (shared_access != IR::SharedAccess::None) {
 				const auto index = inst.Flags<IR::MemoryFlags>().index;
-				if (index >= program.values->memory_info.size()) {
-					return Fail(error, "shared operation has invalid memory metadata");
+				if (index >= program.memory_info.size()) {
+					Fail(program, "shared operation has invalid memory metadata");
 				}
-				const auto kind = program.values->memory_info[index].kind;
+				const auto kind = program.memory_info[index].kind;
 				if (kind != IR::ResourceKind::Lds && kind != IR::ResourceKind::Gds) {
-					return Fail(error, "shared operation has invalid resource kind");
+					Fail(program, "shared operation has invalid resource kind");
 				}
 				if (program.stage != ShaderType::Compute && kind == IR::ResourceKind::Lds) {
 					requirements.function_lds = true;
 				}
 				if (shared_access == IR::SharedAccess::Append ||
 				    shared_access == IR::SharedAccess::Consume) {
-					MarkExactSubgroup();
+					MarkBallot();
 					requirements.subgroup_shuffle             = true;
 					requirements.subgroup_local_invocation_id = true;
 				}
 			}
 			switch (inst.GetOpcode()) {
-				case IR::ValueOpcode::Ballot: MarkExactSubgroup(); break;
+				case IR::ValueOpcode::Ballot: MarkBallot(); break;
 				case IR::ValueOpcode::DppMoveU32:
 				case IR::ValueOpcode::ReadFirstLane:
 				case IR::ValueOpcode::ReadLane: {
-					MarkExactSubgroup();
+					MarkBallot();
 					requirements.subgroup_shuffle = true;
 					if (inst.GetOpcode() == IR::ValueOpcode::DppMoveU32) {
 						requirements.subgroup_local_invocation_id = true;
@@ -270,18 +259,19 @@ bool AnalyzeProgramRequirements(IR::Program& program, std::string* error) {
 				case IR::ValueOpcode::DppUpdateU32:
 				case IR::ValueOpcode::WqmMask:
 				case IR::ValueOpcode::WriteLane: {
-					MarkExactSubgroup();
+					MarkBallot();
 					requirements.subgroup_local_invocation_id = true;
 					break;
 				}
 				case IR::ValueOpcode::Permlane16U32: {
-					MarkExactSubgroup();
+					MarkBallot();
 					requirements.subgroup_shuffle             = true;
 					requirements.subgroup_local_invocation_id = true;
 					break;
 				}
-				case IR::ValueOpcode::SwizzleU32: {
-					MarkExactSubgroup();
+				case IR::ValueOpcode::SwizzleU32:
+				case IR::ValueOpcode::BpermuteU32: {
+					MarkBallot();
 					requirements.subgroup_shuffle             = true;
 					requirements.subgroup_local_invocation_id = true;
 					break;
@@ -295,11 +285,11 @@ bool AnalyzeProgramRequirements(IR::Program& program, std::string* error) {
 					break;
 				case IR::ValueOpcode::SetAttribute: {
 					const auto index = inst.Flags<IR::ExportFlags>().index;
-					if (index >= program.values->export_info.size()) {
-						return Fail(error, "attribute export has invalid metadata");
+					if (index >= program.export_info.size()) {
+						Fail(program, "attribute export has invalid metadata");
 					}
 					if (program.stage == ShaderType::Pixel &&
-					    program.values->export_info[index].vm) {
+					    program.export_info[index].vm) {
 						requirements.pixel_valid_mask = true;
 					}
 					break;
@@ -309,46 +299,33 @@ bool AnalyzeProgramRequirements(IR::Program& program, std::string* error) {
 		}
 	}
 	program.spirv_requirements.emplace(requirements);
-	return true;
 }
 
-bool EmitProgram(const IR::Program& program, const IR::ResourceSnapshot& resources,
-                 ShaderStageInputInfo input_info, std::vector<uint32_t>& spirv,
-                 std::string* error) {
+std::vector<uint32_t> EmitProgram(const IR::Program& program,
+                                  const IR::ResourceSnapshot& resources,
+                                  ShaderStageInputInfo input_info) {
 	using namespace Emitter;
 
 	if (program.stage != ShaderType::Compute && program.stage != ShaderType::Vertex &&
 	    program.stage != ShaderType::Pixel) {
-		SetError(error, "binary SPIR-V emitter supports compute, vertex, and pixel shaders");
-		return false;
+		Fail(program, "binary SPIR-V emitter supports compute, vertex, and pixel shaders");
 	}
 	if (!program.srt_plan_complete || !program.resource_tracking_complete ||
 	    !program.shader_info_complete || !program.binding_layout_complete ||
 	    !program.spirv_requirements.has_value()) {
-		SetError(error, "SPIR-V emitter requires a fully planned native shader program");
-		return false;
+		Fail(program, "SPIR-V emitter requires a fully planned native shader program");
 	}
-	if (!IR::ValidateResourceSnapshot(program, resources, error)) {
-		return false;
+	if (!IR::ValidateResourceSnapshot(program, resources)) {
+		Fail(program, "resource snapshot is inconsistent with the native shader plan");
 	}
-	if (!IR::ValidateResourceSpecialization(program, resources, error)) {
-		return false;
-	}
-	if (!ValidateNativeProgram(program, error)) {
-		return false;
-	}
-	if (program.values == nullptr) {
-		SetError(error, "SPIR-V emitter requires planned typed SSA");
-		return false;
-	}
-	const auto& value_program = *program.values;
-	if (!IR::ValidateValueProgram(value_program, true, error)) {
-		return false;
+	if (!IR::ValidateResourceSpecialization(program, resources)) {
+		Fail(program, "resource specialization is inconsistent with the native shader plan");
 	}
+	ValidateNativeProgram(program);
+	IR::ValidateProgram(program, true);
 	EmitterState state(program, resources, input_info);
-	state.stage                = program.stage;
-	state.wave_size            = program.wave_size;
-	state.per_invocation_masks = program.lane_mask_mode == ShaderLaneMaskMode::PerInvocation;
+	state.stage     = program.stage;
+	state.wave_size = program.wave_size;
 	state.inputs.reserve(program.info.inputs.size());
 	state.outputs.reserve(program.info.outputs.size());
 	state.interface_variables.reserve(program.info.inputs.size() + program.info.outputs.size());
@@ -356,17 +333,13 @@ bool EmitProgram(const IR::Program& program, const IR::ResourceSnapshot& resourc
 	AllocateInputVariables(state);
 	AllocateOutputVariables(state);
 	DefineModule(state);
-	if (!EmitValueProgram(state, value_program, error)) {
-		return false;
-	}
+	EmitProgram(state, program);
 
 	auto binary = state.builder.Build();
 	if (binary.empty()) {
-		SetError(error, "SPIR-V builder returned an empty module");
-		return false;
+		Fail(program, "SPIR-V builder returned an empty module");
 	}
-	spirv = std::move(binary);
-	return true;
+	return binary;
 }
 
 } // namespace Libs::Graphics::ShaderRecompiler::Spirv

src/graphics/shader/recompiler/backend/spirv/SpirvEmitter.h

@@ -9,10 +9,11 @@
 
 namespace Libs::Graphics::ShaderRecompiler::Spirv {
 
-bool AnalyzeProgramRequirements(IR::Program& program, std::string* error);
+void AnalyzeProgramRequirements(IR::Program& program);
 
-bool EmitProgram(const IR::Program& program, const IR::ResourceSnapshot& resources,
-                 ShaderStageInputInfo input_info, std::vector<uint32_t>& spirv, std::string* error);
+std::vector<uint32_t> EmitProgram(const IR::Program& program,
+                                  const IR::ResourceSnapshot& resources,
+                                  ShaderStageInputInfo input_info);
 
 } // namespace Libs::Graphics::ShaderRecompiler::Spirv
 

src/graphics/shader/recompiler/backend/spirv/spirvEmitterAluHelpers.cpp

@@ -223,12 +223,8 @@ F32Class EmitClassifyF32(EmitterState& state, uint32_t value) {
 	const auto mantissa_bits = EmitAndConstant(state, abs_bits, 0x007fffffu);
 	const auto exponent_max  = EmitCompareU32Constant(state, OpIEqual, exponent_bits, 0x7f800000u);
 	const auto mantissa_nonzero = EmitCompareU32Constant(state, OpINotEqual, mantissa_bits, 0);
-	const auto quiet_bit_clear =
-	    EmitCompareU32Constant(state, OpIEqual, EmitAndConstant(state, cls.bits, 0x00400000u), 0);
-	cls.nan        = EmitLogicalAndBool(state, exponent_max, mantissa_nonzero);
-	cls.snan       = EmitLogicalAndBool(state, cls.nan, quiet_bit_clear);
-	cls.zero       = EmitCompareU32Constant(state, OpIEqual, abs_bits, 0);
-	cls.quiet_bits = EmitOrU32(state, cls.bits, ConstantU32(state, 0x00400000u));
+	cls.nan  = EmitLogicalAndBool(state, exponent_max, mantissa_nonzero);
+	cls.zero = EmitCompareU32Constant(state, OpIEqual, abs_bits, 0);
 	return cls;
 }
 
@@ -308,12 +304,7 @@ uint32_t EmitMinMaxF32Value(EmitterState& state, uint32_t lhs, uint32_t rhs, boo
 
 	const auto rhs_nan_bits =
 	    EmitSelectValueU32(state, rhs_class.nan, lhs_class.bits, numeric_bits);
-	const auto non_snan_bits =
-	    EmitSelectValueU32(state, lhs_class.nan, rhs_class.bits, rhs_nan_bits);
-	const auto rhs_snan_bits =
-	    EmitSelectValueU32(state, rhs_class.snan, rhs_class.quiet_bits, non_snan_bits);
-	const auto result_bits =
-	    EmitSelectValueU32(state, lhs_class.snan, lhs_class.quiet_bits, rhs_snan_bits);
+	const auto result_bits = EmitSelectValueU32(state, lhs_class.nan, rhs_class.bits, rhs_nan_bits);
 	return EmitBitcastU32ToF32(state, result_bits);
 }
 

src/graphics/shader/recompiler/backend/spirv/spirvEmitterAnalysis.cpp

@@ -2,7 +2,7 @@
 #include "graphics/guest_gpu/gpu_defs.h"
 #include "graphics/shader/recompiler/backend/spirv/SpirvEmitter.h"
 #include "graphics/shader/recompiler/backend/spirv/spirvEmitterInternal.h"
-#include "graphics/shader/recompiler/ir/ValueProgram.h"
+#include "graphics/shader/recompiler/ir/ShaderIR.h"
 
 namespace Libs::Graphics::ShaderRecompiler::Spirv::Emitter {
 
@@ -32,10 +32,9 @@ bool PixelParameterIsFlat(const EmitterState& state, uint32_t attr) {
 	       ShaderPixelParameterIsFlat(*state.input_info.pixel, attr);
 }
 
-void SetError(std::string* error, const char* message) {
-	if (error != nullptr) {
-		*error = message;
-	}
+bool PixelParameterIsCustom(const EmitterState& state, uint32_t attr) {
+	return state.stage == ShaderType::Pixel &&
+	       ShaderPixelParameterIsCustom(*state.input_info.pixel, attr);
 }
 
 bool HasOutput(const std::vector<OutputBinding>& outputs, IR::StageOutputKind kind,
@@ -47,8 +46,8 @@ bool HasOutput(const std::vector<OutputBinding>& outputs, IR::StageOutputKind ki
 
 void CopyProgramInputsAndOutputs(EmitterState& state, const IR::Program& program) {
 	for (const auto& input: program.info.inputs) {
-		state.inputs.push_back(
-		    {input.kind, input.location, input.component_count, 0, input.debug_name});
+		state.inputs.push_back({input.kind, input.location, input.component_count, 0,
+		                        input.debug_name, input.per_vertex});
 	}
 	for (const auto& output: program.info.outputs) {
 		if (HasOutput(state.outputs, output.kind, output.index)) {
@@ -59,7 +58,7 @@ void CopyProgramInputsAndOutputs(EmitterState& state, const IR::Program& program
 }
 
 uint32_t OutputVariableForExport(const EmitterState& state, const IR::ExportInfo& exp) {
-	if (exp.kind == IR::ExportTargetKind::Position) {
+	if (exp.kind == IR::ExportTargetKind::Position && exp.index == 0) {
 		return state.per_vertex_variable;
 	}
 	if (exp.kind == IR::ExportTargetKind::MrtZ) {

src/graphics/shader/recompiler/backend/spirv/spirvEmitterFlow.cpp

@@ -5,13 +5,11 @@
 namespace Libs::Graphics::ShaderRecompiler::Spirv::Emitter {
 namespace {
 
-bool UserDataDwordIndex(const EmitterState& state, IR::Register reg, uint32_t& dword_index) {
-	if (reg.file != IR::RegisterFile::Scalar) {
-		return false;
-	}
+bool UserDataDwordIndex(const EmitterState& state, IR::ScalarReg reg, uint32_t& dword_index) {
+	const auto register_index = IR::RegIndex(reg);
 	const auto& registers = state.program.bindings.user_data_registers;
-	const auto  found     = std::lower_bound(registers.begin(), registers.end(), reg.index);
-	if (found == registers.end() || *found != reg.index) {
+	const auto  found     = std::lower_bound(registers.begin(), registers.end(), register_index);
+	if (found == registers.end() || *found != register_index) {
 		return false;
 	}
 	dword_index = static_cast<uint32_t>(found - registers.begin());
@@ -58,13 +56,6 @@ uint32_t EmitWqmU64(EmitterState& state, uint32_t value) {
 	return result;
 }
 
-uint32_t BoolAsU32(ValueEmitContext& ctx, uint32_t value) {
-	const auto result = ctx.state.builder.AllocateId();
-	ctx.state.builder.AddFunction({OpSelect, TypeU32(ctx.state), result, value,
-	                               ConstantU32(ctx.state, 1), ConstantU32(ctx.state, 0)});
-	return result;
-}
-
 uint32_t EmitBuiltinU32(ValueEmitContext& ctx, IR::StageInputKind kind, uint32_t component) {
 	auto& state = ctx.state;
 	if (kind == IR::StageInputKind::LocalInvocationIndex) {
@@ -100,6 +91,18 @@ uint32_t EmitBuiltinU32(ValueEmitContext& ctx, IR::StageInputKind kind, uint32_t
 		state.builder.AddFunction({OpBitcast, TypeU32(state), bits, value});
 		return bits;
 	}
+	if (kind == IR::StageInputKind::BaryCoordSmooth ||
+	    kind == IR::StageInputKind::BaryCoordNoPerspective) {
+		const auto pointer = state.builder.AllocateId();
+		const auto value   = state.builder.AllocateId();
+		const auto bits    = state.builder.AllocateId();
+		state.builder.AddFunction({OpAccessChain,
+		                           TypePointer(state, StorageClassInput, TypeF32(state)), pointer,
+		                           variable, ConstantU32(state, component + 1u)});
+		state.builder.AddFunction({OpLoad, TypeF32(state), value, pointer});
+		state.builder.AddFunction({OpBitcast, TypeU32(state), bits, value});
+		return bits;
+	}
 	return EmitInputComponentU32(state, kind, component);
 }
 
@@ -194,6 +197,43 @@ uint32_t EmitAttribute(ValueEmitContext& ctx, uint32_t attr, uint32_t chan) {
 	if (state.stage == ShaderType::Vertex) {
 		return EmitVertexParameterComponentU32(state, *input, chan & 3u);
 	}
+	const auto load_per_vertex = [&](uint32_t vertex) {
+		const auto pointer = state.builder.AllocateId();
+		const auto value   = state.builder.AllocateId();
+		state.builder.AddFunction(
+		    {OpAccessChain, TypePointer(state, StorageClassInput, TypeF32(state)), pointer,
+		     input->variable_id, ConstantU32(state, vertex), ConstantU32(state, chan & 3u)});
+		state.builder.AddFunction({OpLoad, TypeF32(state), value, pointer});
+		return value;
+	};
+	if (input->per_vertex) {
+		const auto barycentric_kind = state.input_info.pixel->ps_no_perspective
+		                                  ? IR::StageInputKind::BaryCoordNoPerspective
+		                                  : IR::StageInputKind::BaryCoordSmooth;
+		const auto barycentric      = InputVariableForKind(state, barycentric_kind);
+		uint32_t   sum              = 0;
+		for (uint32_t vertex = 0; vertex < 3u; vertex++) {
+			const auto pointer = state.builder.AllocateId();
+			const auto weight  = state.builder.AllocateId();
+			const auto product = state.builder.AllocateId();
+			state.builder.AddFunction({OpAccessChain,
+			                           TypePointer(state, StorageClassInput, TypeF32(state)),
+			                           pointer, barycentric, ConstantU32(state, vertex)});
+			state.builder.AddFunction({OpLoad, TypeF32(state), weight, pointer});
+			state.builder.AddFunction(
+			    {OpFMul, TypeF32(state), product, load_per_vertex(vertex), weight});
+			if (vertex == 0u) {
+				sum = product;
+			} else {
+				const auto next = state.builder.AllocateId();
+				state.builder.AddFunction({OpFAdd, TypeF32(state), next, sum, product});
+				sum = next;
+			}
+		}
+		const auto bits = state.builder.AllocateId();
+		state.builder.AddFunction({OpBitcast, TypeU32(state), bits, sum});
+		return bits;
+	}
 	const auto vector    = state.builder.AllocateId();
 	const auto component = state.builder.AllocateId();
 	const auto bits      = state.builder.AllocateId();
@@ -203,10 +243,41 @@ uint32_t EmitAttribute(ValueEmitContext& ctx, uint32_t attr, uint32_t chan) {
 	return bits;
 }
 
-bool MrtUsesUint(const EmitterState& state, const IR::ExportInfo& exp) {
-	return state.stage == ShaderType::Pixel && exp.kind == IR::ExportTargetKind::Mrt &&
-	       exp.index < std::size(state.input_info.pixel->target_output_mode) &&
-	       state.input_info.pixel->target_output_mode[exp.index] == 7u;
+uint32_t EmitInterpolationParameter(ValueEmitContext& ctx, uint32_t attr, uint32_t chan,
+                                    uint32_t mode) {
+	auto&       state = ctx.state;
+	const auto* input = InputBindingForParameter(state, attr);
+	if (!input->per_vertex) {
+		return EmitAttribute(ctx, attr, chan);
+	}
+	const auto load_vertex = [&](uint32_t vertex) {
+		const auto pointer = state.builder.AllocateId();
+		const auto value   = state.builder.AllocateId();
+		state.builder.AddFunction(
+		    {OpAccessChain, TypePointer(state, StorageClassInput, TypeF32(state)), pointer,
+		     input->variable_id, ConstantU32(state, vertex), ConstantU32(state, chan & 3u)});
+		state.builder.AddFunction({OpLoad, TypeF32(state), value, pointer});
+		return value;
+	};
+
+	const auto selected_vertex = (mode + 1u) % 3u;
+	uint32_t   value           = load_vertex(selected_vertex);
+	if (!PixelParameterIsCustom(state, attr) && mode < 2u) {
+		const auto delta = state.builder.AllocateId();
+		state.builder.AddFunction({OpFSub, TypeF32(state), delta, value, load_vertex(0)});
+		value = delta;
+	}
+	const auto bits = state.builder.AllocateId();
+	state.builder.AddFunction({OpBitcast, TypeU32(state), bits, value});
+	return bits;
+}
+
+uint32_t MrtOutputMode(const EmitterState& state, const IR::ExportInfo& exp) {
+	if (state.stage != ShaderType::Pixel || exp.kind != IR::ExportTargetKind::Mrt ||
+	    exp.index >= std::size(state.input_info.pixel->target_output_mode)) {
+		return 0;
+	}
+	return state.input_info.pixel->target_output_mode[exp.index];
 }
 
 uint32_t ExportRawComponent(ValueEmitContext& ctx, uint32_t vector, uint32_t component) {
@@ -220,6 +291,8 @@ uint32_t ExportVector(ValueEmitContext& ctx, uint32_t data, const IR::ExportInfo
                       bool uint_output) {
 	auto& state = ctx.state;
 	if (exp.compr && !uint_output) {
+		const auto unpack =
+		    MrtOutputMode(state, exp) == 5u ? GlslUnpackUnorm2x16 : GlslUnpackHalf2x16;
 		uint32_t f32[4] = {ConstantF32(state, 0), ConstantF32(state, 0), ConstantF32(state, 0),
 		                   ConstantF32(state, 0x3f800000u)};
 		for (uint32_t pair = 0; pair < 2u; pair++) {
@@ -229,8 +302,8 @@ uint32_t ExportVector(ValueEmitContext& ctx, uint32_t data, const IR::ExportInfo
 			const auto packed   = state.builder.AllocateId();
 			const auto unpacked = state.builder.AllocateId();
 			state.builder.AddFunction({OpCompositeExtract, TypeU32(state), packed, data, pair});
-			state.builder.AddFunction({OpExtInst, TypeF32Vector(state, 2), unpacked,
-			                           GlslStd450(state), GlslUnpackHalf2x16, packed});
+			state.builder.AddFunction(
+			    {OpExtInst, TypeF32Vector(state, 2), unpacked, GlslStd450(state), unpack, packed});
 			for (uint32_t lane = 0; lane < 2u; lane++) {
 				const auto component = pair * 2u + lane;
 				if (((exp.en >> component) & 1u) != 0u) {
@@ -292,6 +365,49 @@ uint32_t ExportVector(ValueEmitContext& ctx, uint32_t data, const IR::ExportInfo
 	return vector;
 }
 
+void EmitAuxPositionExport(ValueEmitContext& ctx, uint32_t data, const IR::ExportInfo& exp) {
+	auto& state = ctx.state;
+	for (uint32_t component = 0; component < 4; component++) {
+		if ((exp.en & (1u << component)) == 0) {
+			continue;
+		}
+		const auto output = IR::DecodePositionExportComponent(
+		    state.input_info.vertex->pa_cl_vs_out_cntl, exp.index, component);
+		if (output.layer) {
+			const auto raw   = ExportRawComponent(ctx, data, component);
+			const auto layer = state.builder.AllocateId();
+			state.builder.AddFunction({OpBitwiseAnd, TypeU32(state), layer, raw,
+			                           ConstantU32(state, 0x7ffu)});
+			state.builder.AddFunction({OpStore, state.layer_variable, layer});
+			continue;
+		}
+		if (!output.point_size && output.clip_distance == UINT32_MAX &&
+		    output.cull_distance == UINT32_MAX) {
+			continue;
+		}
+
+		const auto raw = ExportRawComponent(ctx, data, component);
+		const auto f32 = state.builder.AllocateId();
+		state.builder.AddFunction({OpBitcast, TypeF32(state), f32, raw});
+		if (output.point_size) {
+			state.builder.AddFunction({OpStore, state.point_size_variable, f32});
+			continue;
+		}
+		auto StoreDistance = [&](uint32_t variable, uint32_t index) {
+			if (index == UINT32_MAX) {
+				return;
+			}
+			const auto pointer = state.builder.AllocateId();
+			state.builder.AddFunction(
+			    {OpAccessChain, TypePointer(state, StorageClassOutput, TypeF32(state)), pointer,
+			     variable, ConstantU32(state, index)});
+			state.builder.AddFunction({OpStore, pointer, f32});
+		};
+		StoreDistance(state.clip_distance_variable, output.clip_distance);
+		StoreDistance(state.cull_distance_variable, output.cull_distance);
+	}
+}
+
 void EmitExport(ValueEmitContext& ctx, const IR::Inst& inst) {
 	auto&       state = ctx.state;
 	const auto& exp   = ctx.Export(inst);
@@ -309,6 +425,10 @@ void EmitExport(ValueEmitContext& ctx, const IR::Inst& inst) {
 	}
 	EmitIfCondition(state, exec, [&]() {
 		const auto data = ctx.Arg(inst, 0);
+		if (exp.kind == IR::ExportTargetKind::Position && exp.index != 0) {
+			EmitAuxPositionExport(ctx, data, exp);
+			return;
+		}
 		if (exp.kind == IR::ExportTargetKind::MrtZ) {
 			if ((exp.en & 1u) != 0u && state.depth_variable != 0) {
 				const auto raw = ExportRawComponent(ctx, data, 0);
@@ -332,7 +452,7 @@ void EmitExport(ValueEmitContext& ctx, const IR::Inst& inst) {
 		if (variable == 0) {
 			return;
 		}
-		const bool uint_output = MrtUsesUint(state, exp);
+		const bool uint_output = MrtOutputMode(state, exp) == 7u;
 		const auto vector_type = uint_output ? TypeU32Vector(state, 4) : TypeF32Vector(state, 4);
 		auto       value       = ExportVector(ctx, data, exp, uint_output);
 		if (state.stage == ShaderType::Pixel && exp.kind == IR::ExportTargetKind::Mrt &&
@@ -382,7 +502,7 @@ bool EmitValueFlow(ValueEmitContext& ctx, const IR::Inst& inst) {
 		case IR::ValueOpcode::GetUserData: {
 			const auto reg   = inst.Arg(0).ScalarRegister();
 			uint32_t   dword = 0;
-			if (!UserDataDwordIndex(state, {IR::RegisterFile::Scalar, IR::RegIndex(reg)}, dword)) {
+			if (!UserDataDwordIndex(state, reg, dword)) {
 				ctx.Define(inst, ConstantU32(state, 0));
 			} else {
 				ctx.Define(inst, EmitShaderDataDwordLoad(state, dword));
@@ -437,16 +557,8 @@ bool EmitValueFlow(ValueEmitContext& ctx, const IR::Inst& inst) {
 			ctx.Define(inst, EmitSubgroupLocalInvocationId(state));
 			return true;
 		case IR::ValueOpcode::Ballot:
-			if (state.per_invocation_masks) {
-				const auto result = state.builder.AllocateId();
-				state.builder.AddFunction({OpCompositeConstruct, TypeU32Vector(state, 4), result,
-				                           BoolAsU32(ctx, ctx.Arg(inst, 0)), ConstantU32(state, 0),
-				                           ConstantU32(state, 0), ConstantU32(state, 0)});
-				ctx.Define(inst, result);
-			} else {
-				ctx.Emit(inst, OpGroupNonUniformBallot, IR::Type::U32x4,
-				         {ConstantU32(state, ScopeSubgroup), ctx.Arg(inst, 0)});
-			}
+			ctx.Emit(inst, OpGroupNonUniformBallot, IR::Type::U32x4,
+			         {ConstantU32(state, ScopeSubgroup), ctx.Arg(inst, 0)});
 			return true;
 		case IR::ValueOpcode::ReadFirstLane: {
 			const auto ballot = state.builder.AllocateId();
@@ -526,6 +638,10 @@ bool EmitValueFlow(ValueEmitContext& ctx, const IR::Inst& inst) {
 		case IR::ValueOpcode::GetAttribute:
 			ctx.Define(inst, EmitAttribute(ctx, inst.Arg(0).U32(), inst.Arg(1).U32()));
 			return true;
+		case IR::ValueOpcode::GetInterpolationParameter:
+			ctx.Define(inst, EmitInterpolationParameter(ctx, inst.Arg(0).U32(), inst.Arg(1).U32(),
+			                                            inst.Arg(2).U32()));
+			return true;
 		case IR::ValueOpcode::SetAttribute: EmitExport(ctx, inst); return true;
 		case IR::ValueOpcode::GetShaderBase:
 			// Guest S_GETPC values stay shader-relative in SPIR-V, matching the runtime ABI. The

src/graphics/shader/recompiler/backend/spirv/spirvEmitterImage.cpp

@@ -439,6 +439,49 @@ uint32_t UnpackImageGather(ValueEmitContext& ctx, const IR::MemoryInfo& mem, uin
 	return result;
 }
 
+uint32_t EmitOneDimensionalGatherLz(ValueEmitContext& ctx, const IR::MemoryInfo& mem, uint32_t pc,
+	                                 uint32_t coord, bool integer) {
+	auto& state = ctx.state;
+	state.builder.RequireCapability(CapabilityImageQuery);
+	const auto image = LoadSampledImageDescriptor(state, mem, pc, ImageViewKind::Dim1D);
+	const auto width = state.builder.AllocateId();
+	state.builder.AddFunction(
+	    {OpImageQuerySizeLod, TypeU32(state), width, image, ConstantU32(state, 0)});
+	const auto width_f32 = state.builder.AllocateId();
+	state.builder.AddFunction({OpConvertUToF, TypeF32(state), width_f32, width});
+	const auto left = state.builder.AllocateId();
+	state.builder.AddFunction(
+	    {OpExtInst, TypeF32(state), left, GlslStd450(state), GlslFloor,
+	     Binary(state, OpFSub, TypeF32(state),
+	            Binary(state, OpFMul, TypeF32(state), coord, width_f32),
+	            ConstantF32(state, 0x3f000000u))});
+
+	const auto sampled = MakeSampledImage(state, mem, pc, ImageViewKind::Dim1D);
+	const auto vector_type = integer ? TypeU32Vector(state, 4) : TypeF32Vector(state, 4);
+	const auto scalar_type = integer ? TypeU32(state) : TypeF32(state);
+	const auto component = ImageConversionFormat(state, mem).format == Prospero::BufferFormat::kInvalid
+	                           ? ImageGatherComponent(mem.dmask)
+	                           : 0u;
+	uint32_t values[2] {};
+	for (uint32_t index = 0; index < 2u; index++) {
+		const auto sample_coord = Binary(
+		    state, OpFDiv, TypeF32(state),
+		    Binary(state, OpFAdd, TypeF32(state), left,
+		           ConstantF32(state, index == 0u ? 0x3f000000u : 0x3fc00000u)),
+		    width_f32);
+		const auto texel = state.builder.AllocateId();
+		state.builder.AddFunction({OpImageSampleExplicitLod, vector_type, texel, sampled, sample_coord,
+		                           ImageOperandsLodMask, ZeroF32(state)});
+		values[index] = state.builder.AllocateId();
+		state.builder.AddFunction(
+		    {OpCompositeExtract, scalar_type, values[index], texel, component});
+	}
+	const auto result = state.builder.AllocateId();
+	state.builder.AddFunction({OpCompositeConstruct, vector_type, result, values[0], values[1],
+	                           values[1], values[0]});
+	return result;
+}
+
 uint32_t PackImageTexel(ValueEmitContext& ctx, const IR::MemoryInfo& mem, uint32_t texel) {
 	const auto info = ImageConversionFormat(ctx.state, mem);
 	if (info.format == Prospero::BufferFormat::kInvalid) return texel;
@@ -613,6 +656,22 @@ bool EmitValueImage(ValueEmitContext& ctx, const IR::Inst& inst) {
 		const auto coord =
 		    CoordF32(ctx, mem, *address, layout.coord, ImageViewCoordinateComponents(view));
 		if (op == IR::ValueOpcode::ImageGatherRaw) {
+			if (view == ImageViewKind::Dim1D) {
+				if (dref || !HasFlag(mem, Decoder::ImageSampleFlagLevelZero) ||
+				    HasFlag(mem, Decoder::ImageSampleFlagOffset) ||
+				    HasFlag(mem, Decoder::ImageSampleFlagGatherHorizontal)) {
+					ctx.Fail(inst, "has an unsupported 1D gather variant");
+					return true;
+				}
+				const auto sample = EmitOneDimensionalGatherLz(ctx, mem, pc, coord, integer);
+				ctx.Define(inst, ResultVector(ctx, UnpackImageGather(ctx, mem, sample), integer,
+				                              false, mem, true));
+				return true;
+			}
+			if (view == ImageViewKind::Dim1DArray) {
+				ctx.Fail(inst, "has an unsupported 1D-array gather");
+				return true;
+			}
 			const auto            sampled = MakeSampledImage(state, mem, pc, view);
 			const auto            sample  = state.builder.AllocateId();
 			std::vector<uint32_t> words =

src/graphics/shader/recompiler/backend/spirv/spirvEmitterInternal.h

@@ -6,7 +6,7 @@
 #include "graphics/shader/recompiler/BufferFormat.h"
 #include "graphics/shader/recompiler/backend/spirv/SpirvBuilder.h"
 #include "graphics/shader/recompiler/ir/ShaderIR.h"
-#include "graphics/shader/recompiler/ir/ValueProgram.h"
+#include "graphics/shader/recompiler/ir/ShaderIR.h"
 #include "graphics/shader/recompiler/ir/passes/BindingLayout.h"
 #include "graphics/shader/recompiler/ir/passes/ResourceMaterialization.h"
 
@@ -36,9 +36,13 @@ enum : uint32_t {
 	ExecutionModeSignedZeroInfNanPreserve    = 4461,
 	ExecutionModeDerivativeGroupQuadsKHR     = 5289,
 	AddressingModelLogical                   = 0,
+	AddressingModelPhysicalStorageBuffer64   = 5348,
 	MemoryModelGLSL450                       = 1,
 	CapabilityShader                         = 1,
+	CapabilityInt64                          = 11,
 	CapabilityImageGatherExtended            = 25,
+	CapabilityClipDistance                   = 32,
+	CapabilityCullDistance                   = 33,
 	CapabilitySampled1D                      = 43,
 	CapabilityImage1D                        = 44,
 	CapabilityImageQuery                     = 50,
@@ -47,7 +51,10 @@ enum : uint32_t {
 	CapabilityGroupNonUniformBallot          = 64,
 	CapabilityGroupNonUniformShuffle         = 65,
 	CapabilitySignedZeroInfNanPreserve       = 4466,
+	CapabilityShaderViewportIndexLayerEXT    = 5254,
+	CapabilityFragmentBarycentricKHR         = 5284,
 	CapabilityComputeDerivativeGroupQuadsKHR = 5288,
+	CapabilityPhysicalStorageBufferAddresses = 5347,
 	StorageClassUniformConstant              = 0,
 	StorageClassInput                        = 1,
 	StorageClassOutput                       = 3,
@@ -56,6 +63,7 @@ enum : uint32_t {
 	StorageClassPushConstant                 = 9,
 	StorageClassImage                        = 11,
 	StorageClassStorageBuffer                = 12,
+	StorageClassPhysicalStorageBuffer        = 5349,
 	FunctionControlNone                      = 0,
 	SelectionControlNone                     = 0,
 	LoopControlNone                          = 0,
@@ -71,10 +79,15 @@ enum : uint32_t {
 	DecorationBinding       = 33,
 	DecorationDescriptorSet = 34,
 	DecorationOffset        = 35,
+	DecorationPerVertexKHR  = 5285,
 };
 
 enum : uint32_t {
 	BuiltInPosition                  = 0,
+	BuiltInPointSize                 = 1,
+	BuiltInClipDistance              = 3,
+	BuiltInCullDistance              = 4,
+	BuiltInLayer                     = 9,
 	BuiltInFragCoord                 = 15,
 	BuiltInFrontFacing               = 17,
 	BuiltInSampleMask                = 20,
@@ -86,6 +99,8 @@ enum : uint32_t {
 	BuiltInSubgroupLocalInvocationId = 41,
 	BuiltInVertexIndex               = 42,
 	BuiltInInstanceIndex             = 43,
+	BuiltInBaryCoordKHR              = 5286,
+	BuiltInBaryCoordNoPerspKHR       = 5287,
 };
 
 enum : uint32_t {
@@ -137,7 +152,9 @@ enum : uint32_t {
 	OpConstantComposite            = 44,
 	OpUndef                        = 1,
 	OpFunction                     = 54,
+	OpFunctionParameter            = 55,
 	OpFunctionEnd                  = 56,
+	OpFunctionCall                 = 57,
 	OpVariable                     = 59,
 	OpImageTexelPointer            = 60,
 	OpLoad                         = 61,
@@ -167,6 +184,8 @@ enum : uint32_t {
 	OpConvertFToS                  = 110,
 	OpConvertSToF                  = 111,
 	OpConvertUToF                  = 112,
+	OpUConvert                     = 113,
+	OpConvertUToPtr                = 120,
 	OpBitcast                      = 124,
 	OpSNegate                      = 126,
 	OpFNegate                      = 127,
@@ -245,35 +264,41 @@ enum : uint32_t {
 	OpSwitch                       = 251,
 	OpKill                         = 252,
 	OpReturn                       = 253,
+	OpReturnValue                  = 254,
 	OpGroupNonUniformBallot        = 339,
 	OpGroupNonUniformBallotFindLSB = 343,
 	OpGroupNonUniformShuffle       = 345,
 };
 
 enum : uint32_t {
-	GlslRoundEven      = 2,
-	GlslTrunc          = 3,
-	GlslFAbs           = 4,
-	GlslFloor          = 8,
-	GlslCeil           = 9,
-	GlslFract          = 10,
-	GlslSin            = 13,
-	GlslCos            = 14,
-	GlslExp2           = 29,
-	GlslLog2           = 30,
-	GlslSqrt           = 31,
-	GlslInverseSqrt    = 32,
-	GlslFMin           = 37,
-	GlslFMax           = 40,
-	GlslFClamp         = 43,
-	GlslLdexp          = 53,
-	GlslFma            = 50,
-	GlslPackSnorm2x16  = 56,
-	GlslPackUnorm2x16  = 57,
-	GlslPackHalf2x16   = 58,
-	GlslUnpackHalf2x16 = 62,
-	GlslFindILsb       = 73,
-	GlslFindUMsb       = 75,
+	MemoryAccessAlignedMask = 0x2,
+};
+
+enum : uint32_t {
+	GlslRoundEven       = 2,
+	GlslTrunc           = 3,
+	GlslFAbs            = 4,
+	GlslFloor           = 8,
+	GlslCeil            = 9,
+	GlslFract           = 10,
+	GlslSin             = 13,
+	GlslCos             = 14,
+	GlslExp2            = 29,
+	GlslLog2            = 30,
+	GlslSqrt            = 31,
+	GlslInverseSqrt     = 32,
+	GlslFMin            = 37,
+	GlslFMax            = 40,
+	GlslFClamp          = 43,
+	GlslLdexp           = 53,
+	GlslFma             = 50,
+	GlslPackSnorm2x16   = 56,
+	GlslPackUnorm2x16   = 57,
+	GlslPackHalf2x16    = 58,
+	GlslUnpackUnorm2x16 = 61,
+	GlslUnpackHalf2x16  = 62,
+	GlslFindILsb        = 73,
+	GlslFindUMsb        = 75,
 };
 
 struct InputBinding {
@@ -282,6 +307,7 @@ struct InputBinding {
 	uint32_t           component_count = 1;
 	uint32_t           variable_id     = 0;
 	std::string        debug_name;
+	bool               per_vertex = false;
 };
 
 struct OutputBinding {
@@ -324,41 +350,47 @@ struct EmitterState {
 	    : program(program_), resources(resources_), input_info(input_info_),
 	      requirements(*program_.spirv_requirements) {}
 
-	Builder                                          builder;
-	const IR::Program&                               program;
-	const IR::ResourceSnapshot&                      resources;
-	ShaderStageInputInfo                             input_info;
-	const IR::SpirvRequirements&                     requirements;
-	ShaderType                                       stage                   = ShaderType::Unknown;
-	uint32_t                                         wave_size               = 64;
-	bool                                             per_invocation_masks    = false;
-	uint32_t                                         storage_buffer_variable = 0;
-	std::array<uint32_t, IR::ShaderInfo::MaxBuffers + IR::ShaderInfo::MaxAddresses>
-	                                                 memory_byte_offsets {};
-	uint32_t                                         address_memory_variable = 0;
-	uint32_t                                         gds_variable            = 0;
-	uint32_t                                         gds_length              = 0;
-	uint32_t                                         push_constant_variable  = 0;
-	uint32_t                                         vsharp_storage_variable = 0;
-	uint32_t                                         flattened_srt_variable  = 0;
-	uint32_t                                         lds_variable            = 0;
-	uint32_t                                         scratch_variable        = 0;
+	Builder                      builder;
+	const IR::Program&           program;
+	const IR::ResourceSnapshot&  resources;
+	ShaderStageInputInfo         input_info;
+	const IR::SpirvRequirements& requirements;
+	ShaderType                   stage                   = ShaderType::Unknown;
+	uint32_t                     wave_size               = 64;
+	uint32_t                     storage_buffer_variable = 0;
+	std::array<uint32_t, IR::ShaderInfo::MaxBuffers>      memory_byte_offsets {};
+	uint32_t                                             bda_pagetable_variable = 0;
+	uint32_t                                             fault_buffer_variable  = 0;
+	uint32_t                                             bda_pointer_function   = 0;
+	uint32_t                                             gds_variable            = 0;
+	uint32_t                                             gds_length              = 0;
+	uint32_t                                             push_constant_variable  = 0;
+	uint32_t                                             vsharp_storage_variable = 0;
+	uint32_t                                             flattened_srt_variable  = 0;
+	uint32_t                                             lds_variable            = 0;
+	uint32_t                                             scratch_variable        = 0;
 	std::array<uint32_t, 2u * SampledImageViewKindCount> sampled_image_variables {};
 	std::array<uint32_t,
 	           static_cast<uint32_t>(StorageImageClass::Count) * StorageImageViewKindCount>
-	                                                 storage_image_variables {};
-	uint32_t                                         sampler_variable                      = 0;
-	uint32_t                                         main_func                             = 0;
-	uint32_t                                         entry_label                           = 0;
-	uint32_t                                         current_label                         = 0;
-	uint32_t                                         pixel_valid_mask_variable             = 0;
-	uint32_t                                         subgroup_local_invocation_id_variable = 0;
-	uint32_t                                         per_vertex_variable                   = 0;
-	uint32_t                                         depth_variable                        = 0;
-	uint32_t                                         sample_mask_variable                  = 0;
-	std::vector<InputBinding>                        inputs;
-	std::vector<OutputBinding>                       outputs;
-	std::vector<uint32_t>                            interface_variables;
+	                           storage_image_variables {};
+	uint32_t                   sampler_variable                      = 0;
+	uint32_t                   main_func                             = 0;
+	uint32_t                   entry_label                           = 0;
+	uint32_t                   current_label                         = 0;
+	uint32_t                   pixel_valid_mask_variable             = 0;
+	uint32_t                   subgroup_local_invocation_id_variable = 0;
+	uint32_t                   per_vertex_variable                   = 0;
+	uint32_t                   point_size_variable                   = 0;
+	uint32_t                   clip_distance_variable                = 0;
+	uint32_t                   cull_distance_variable                = 0;
+	uint32_t                   layer_variable                        = 0;
+	uint32_t                   clip_distance_count                   = 0;
+	uint32_t                   cull_distance_count                   = 0;
+	uint32_t                   depth_variable                        = 0;
+	uint32_t                   sample_mask_variable                  = 0;
+	std::vector<InputBinding>  inputs;
+	std::vector<OutputBinding> outputs;
+	std::vector<uint32_t>      interface_variables;
 };
 
 uint32_t TypeVoid(EmitterState& state);
@@ -366,6 +398,7 @@ uint32_t TypeBool(EmitterState& state);
 uint32_t TypeBoolVector(EmitterState& state, uint32_t components);
 uint32_t TypeU32(EmitterState& state);
 uint32_t TypeU64(EmitterState& state);
+uint32_t TypeDeviceAddress(EmitterState& state);
 uint32_t TypeU32Pair(EmitterState& state);
 uint32_t TypeI32(EmitterState& state);
 uint32_t TypeI32Pair(EmitterState& state);
@@ -379,6 +412,8 @@ uint32_t TypePointer(EmitterState& state, uint32_t storage_class, uint32_t point
 uint32_t TypeFunction(EmitterState& state);
 uint32_t TypeStorageBufferPointer(EmitterState& state);
 uint32_t TypeStorageBufferElementPointer(EmitterState& state);
+uint32_t TypeDeviceAddressStoragePointer(EmitterState& state);
+uint32_t TypePhysicalU32Pointer(EmitterState& state);
 uint32_t TypePushConstantElementPointer(EmitterState& state);
 uint32_t TypeU32ArrayPointer(EmitterState& state, uint32_t storage_class, uint32_t dwords);
 uint32_t TypeU32ElementPointer(EmitterState& state, uint32_t storage_class);
@@ -389,7 +424,7 @@ inline void EmitLabel(EmitterState& state, uint32_t label) {
 }
 
 struct ValueEmitContext {
-	ValueEmitContext(EmitterState& state_, const IR::ValueProgram& program_)
+	ValueEmitContext(EmitterState& state_, const IR::Program& program_)
 	    : state(state_), program(program_) {}
 
 	uint32_t              Def(IR::Value value);
@@ -404,18 +439,17 @@ struct ValueEmitContext {
 	const IR::MemoryInfo& Memory(const IR::Inst& inst) const;
 	const IR::ExportInfo& Export(const IR::Inst& inst) const;
 	uint32_t              Label(const IR::Block* block) const;
-	void                  Fail(const IR::Inst& inst, const char* reason);
+	[[noreturn]] void     Fail(const char* reason) const;
+	[[noreturn]] void     Fail(const IR::Inst& inst, const char* reason) const;
 
 	EmitterState&                                                      state;
-	const IR::ValueProgram&                                            program;
+	const IR::Program&                                            program;
 	std::unordered_map<const IR::Inst*, uint32_t>                      definitions;
 	std::unordered_map<const IR::Block*, uint32_t>                     labels;
 	const std::unordered_map<const IR::Inst*, uint32_t>*               dispatcher_spills = nullptr;
 	std::unordered_map<const IR::Inst*, std::pair<uint32_t, uint32_t>> dispatcher_block_loads;
 	const IR::Block*                                                   current_block = nullptr;
 	uint32_t                                                           scratch_u32_variable = 0;
-	bool                                                               failed               = false;
-	std::string                                                        error;
 };
 
 enum class VertexInputScalarKind { Float, Sint, Uint };
@@ -551,11 +585,9 @@ constexpr uint32_t ImageSpirvMultisampled(ImageViewKind view) {
 }
 
 struct F32Class {
-	uint32_t bits       = 0;
-	uint32_t nan        = 0;
-	uint32_t snan       = 0;
-	uint32_t zero       = 0;
-	uint32_t quiet_bits = 0;
+	uint32_t bits = 0;
+	uint32_t nan  = 0;
+	uint32_t zero = 0;
 };
 
 uint32_t PixelParameterMappedLocation(const EmitterState& state, uint32_t attr);
@@ -564,6 +596,8 @@ uint32_t PixelParameterLocation(const EmitterState& state, uint32_t attr);
 
 bool PixelParameterIsFlat(const EmitterState& state, uint32_t attr);
 
+bool PixelParameterIsCustom(const EmitterState& state, uint32_t attr);
+
 VertexInputScalarKind VertexParameterScalarKind(const EmitterState& state, uint32_t location);
 
 uint32_t VertexParameterComponentCount(const EmitterState& state, const InputBinding& input);
@@ -578,8 +612,6 @@ uint32_t VertexParameterVectorOrScalarType(EmitterState& state, VertexInputScala
 uint32_t VertexParameterInputPointerType(EmitterState& state, VertexInputScalarKind kind,
                                          uint32_t components);
 
-void SetError(std::string* error, const char* message);
-
 bool HasOutput(const std::vector<OutputBinding>& outputs, IR::StageOutputKind kind, uint32_t index);
 
 void CopyProgramInputsAndOutputs(EmitterState& state, const IR::Program& program);
@@ -667,8 +699,8 @@ void AddInputAnnotationsAndNames(EmitterState& state);
 
 void AddOutputAnnotationsAndNames(EmitterState& state);
 
-void DecorateDescriptor(EmitterState& state, uint32_t variable, const char* name, uint32_t set,
-                        uint32_t binding);
+void DecorateDescriptor(EmitterState& state, uint32_t variable, const char* name,
+                        IR::DescriptorBindingKind kind);
 
 void AddDescriptorAnnotationsAndNames(EmitterState& state);
 
@@ -829,7 +861,9 @@ bool EmitValueMemory(ValueEmitContext& ctx, const IR::Inst& inst);
 
 bool EmitValueImage(ValueEmitContext& ctx, const IR::Inst& inst);
 
-bool EmitValueProgram(EmitterState& state, const IR::ValueProgram& program, std::string* error);
+void EmitProgram(EmitterState& state, const IR::Program& program);
+
+void DefineGetBdaPointer(EmitterState& state);
 
 // These templates accept local lambdas from several emitter translation units.
 template <typename Fn>

src/graphics/shader/recompiler/backend/spirv/spirvEmitterMemory.cpp

@@ -1,5 +1,7 @@
 #include "graphics/shader/recompiler/backend/spirv/spirvEmitterInternal.h"
 
+#include "graphics/host_gpu/renderer/cache/bufferCache.h"
+
 #include <algorithm>
 
 namespace Libs::Graphics::ShaderRecompiler::Spirv::Emitter {
@@ -28,6 +30,19 @@ uint32_t AndCondition(EmitterState& state, uint32_t lhs, uint32_t rhs) {
 	return Binary(state, OpLogicalAnd, TypeBool(state), lhs, rhs);
 }
 
+uint32_t EmitDsMaskedLaneRead(EmitterState& state, uint32_t source, uint32_t target,
+                              uint32_t exec) {
+	const auto shuffled = state.builder.AllocateId();
+	state.builder.AddFunction({OpGroupNonUniformShuffle, TypeU32(state), shuffled,
+	                           ConstantU32(state, ScopeSubgroup), source, target});
+	const auto source_exec = state.builder.AllocateId();
+	state.builder.AddFunction({OpGroupNonUniformShuffle, TypeBool(state), source_exec,
+	                           ConstantU32(state, ScopeSubgroup), exec, target});
+	const auto source_active =
+	    AndCondition(state, source_exec, EmitSubgroupLaneActiveBool(state, target));
+	return Select(state, TypeU32(state), source_active, shuffled, ConstantU32(state, 0));
+}
+
 uint32_t BufferByteAddress(ValueEmitContext& ctx, const IR::Inst& inst, const IR::MemoryInfo& mem,
                            uint32_t index, uint32_t offset, uint32_t soffset) {
 	auto&          state   = ctx.state;
@@ -95,60 +110,146 @@ uint32_t AddU64Low(EmitterState& state, uint32_t low, uint32_t high, uint32_t ad
 	return result;
 }
 
-uint32_t AddressByteAddress(ValueEmitContext& ctx, const IR::Inst& inst, const IR::MemoryInfo& mem,
-                            uint32_t low, uint32_t high) {
+uint32_t ScratchByteAddress(ValueEmitContext& ctx, const IR::MemoryInfo& mem, uint32_t low,
+                            uint32_t high) {
 	auto& state     = ctx.state;
 	auto  immediate = static_cast<int32_t>(mem.offset);
-	if (mem.kind == IR::ResourceKind::ScalarAddress) {
-		immediate = static_cast<int32_t>(static_cast<uint32_t>(immediate) & ~3u);
-		low       = Binary(state, OpBitwiseAnd, TypeU32(state), low, ConstantU32(state, ~3u));
-	}
 	const auto immediate_low  = ConstantU32(state, static_cast<uint32_t>(immediate));
 	const auto immediate_high = ConstantU32(state, immediate < 0 ? UINT32_MAX : 0u);
-	if (mem.kind == IR::ResourceKind::Scratch) {
-		low              = AddU64Low(state, low, high, immediate_low, immediate_high, high);
-		const auto valid = Binary(state, OpIEqual, TypeBool(state), high, ConstantU32(state, 0));
-		return Select(state, TypeU32(state), valid, low, ConstantU32(state, UINT32_MAX));
-	}
-	const auto base = state.program.info.addresses[mem.resource].specialized_base;
-	if (mem.kind == IR::ResourceKind::Flat ||
-	    (mem.kind == IR::ResourceKind::Global &&
-	     state.program.info.addresses[mem.resource].unbased)) {
-		low                      = AddU64Low(state, low, high, immediate_low, immediate_high, high);
-		const auto base_low      = ConstantU32(state, static_cast<uint32_t>(base));
-		const auto base_high     = ConstantU32(state, static_cast<uint32_t>(base >> 32u));
-		const auto relative      = Binary(state, OpISub, TypeU32(state), low, base_low);
-		const auto borrow        = Binary(state, OpULessThan, TypeBool(state), low, base_low);
-		const auto relative_high = Binary(
-		    state, OpISub, TypeU32(state), Binary(state, OpISub, TypeU32(state), high, base_high),
-		    Select(state, TypeU32(state), borrow, ConstantU32(state, 1), ConstantU32(state, 0)));
-		const auto valid =
-		    Binary(state, OpIEqual, TypeBool(state), relative_high, ConstantU32(state, 0));
-		return Select(state, TypeU32(state), valid, relative, ConstantU32(state, UINT32_MAX));
-	}
-	const auto initial       = base + static_cast<uint64_t>(static_cast<int64_t>(immediate));
-	auto       relative_high = ConstantU32(state, static_cast<uint32_t>(initial >> 32u));
-	const auto relative      = AddU64Low(state, ConstantU32(state, static_cast<uint32_t>(initial)),
-	                                     relative_high, low, ConstantU32(state, 0), relative_high);
+	low                      = AddU64Low(state, low, high, immediate_low, immediate_high, high);
 	const auto valid =
-	    Binary(state, OpIEqual, TypeBool(state), relative_high, ConstantU32(state, 0));
-	return Select(state, TypeU32(state), valid, relative, ConstantU32(state, UINT32_MAX));
+	    Binary(state, OpIEqual, TypeBool(state), high, ConstantU32(state, 0));
+	return Select(state, TypeU32(state), valid, low, ConstantU32(state, UINT32_MAX));
 }
 
-uint32_t AddAddressMemoryOffset(ValueEmitContext& ctx, const IR::MemoryInfo& mem,
-                                uint32_t address) {
-	auto&      state = ctx.state;
-	const auto binding =
-	    ResourceForDescriptor(state, IR::DescriptorBindingKind::AddressMemory, mem.resource);
-	const auto offset =
-	    state.memory_byte_offsets[state.program.info.buffers.size() + binding.array_index];
-	const auto corrected = Binary(state, OpIAdd, TypeU32(state), address, offset);
-	const auto not_invalid =
-	    Binary(state, OpINotEqual, TypeBool(state), address, ConstantU32(state, UINT32_MAX));
-	const auto no_overflow =
-	    Binary(state, OpUGreaterThanEqual, TypeBool(state), corrected, address);
-	return Select(state, TypeU32(state), AndCondition(state, not_invalid, no_overflow), corrected,
-	              ConstantU32(state, UINT32_MAX));
+uint32_t ConstantDeviceAddress(EmitterState& state, uint64_t value) {
+	return state.builder.Constant(OpConstant, TypeDeviceAddress(state),
+	                              {static_cast<uint32_t>(value),
+	                               static_cast<uint32_t>(value >> 32u)});
+}
+
+uint32_t DeviceAddressFromWords(EmitterState& state, uint32_t low, uint32_t high) {
+	const auto low64  = Unary(state, OpUConvert, TypeDeviceAddress(state), low);
+	const auto high64 = Binary(state, OpShiftLeftLogical, TypeDeviceAddress(state),
+	                           Unary(state, OpUConvert, TypeDeviceAddress(state), high),
+	                           ConstantDeviceAddress(state, 32));
+	return Binary(state, OpBitwiseOr, TypeDeviceAddress(state), low64, high64);
+}
+
+uint32_t GuestAddress(ValueEmitContext& ctx, const IR::Inst& inst, const IR::MemoryInfo& mem) {
+	auto& state = ctx.state;
+	auto  low   = ctx.Arg(inst, 1);
+	if (mem.kind == IR::ResourceKind::ScalarAddress) {
+		low = Binary(state, OpBitwiseAnd, TypeU32(state), low, ConstantU32(state, ~3u));
+	}
+	uint32_t address = 0;
+	if (mem.address_is_full) {
+		address = DeviceAddressFromWords(state, low, ctx.Arg(inst, 2));
+	} else {
+		const auto* handle = inst.Arg(0).Resolve().TryInstruction();
+		if (handle == nullptr || handle->GetOpcode() != IR::ValueOpcode::GetAddressResource ||
+		    handle->NumArgs() != 2) {
+			ctx.Fail(inst, "has no address base pair");
+			return ConstantDeviceAddress(state, 0);
+		}
+		const auto base = DeviceAddressFromWords(state, ctx.Arg(*handle, 0), ctx.Arg(*handle, 1));
+		address = Binary(state, OpIAdd, TypeDeviceAddress(state), base,
+		                 Unary(state, OpUConvert, TypeDeviceAddress(state), low));
+	}
+	auto immediate = static_cast<int32_t>(mem.offset);
+	if (mem.kind == IR::ResourceKind::ScalarAddress) {
+		immediate = static_cast<int32_t>(static_cast<uint32_t>(immediate) & ~3u);
+	}
+	return immediate == 0
+	           ? address
+	           : Binary(state, OpIAdd, TypeDeviceAddress(state), address,
+	                    ConstantDeviceAddress(state,
+	                                          static_cast<uint64_t>(static_cast<int64_t>(immediate))));
+}
+
+uint32_t FaultElementPointer(EmitterState& state, uint32_t index) {
+	const auto pointer = state.builder.AllocateId();
+	state.builder.AddFunction({OpAccessChain, TypeStorageBufferElementPointer(state), pointer,
+	                           state.fault_buffer_variable, ConstantU32(state, 0), index});
+	return pointer;
+}
+
+void RecordBdaFault(EmitterState& state, uint32_t page) {
+	const auto word = Binary(state, OpShiftRightLogical, TypeU32(state), page,
+	                         ConstantU32(state, 5));
+	const auto bit = Binary(
+	    state, OpShiftLeftLogical, TypeU32(state), ConstantU32(state, 1),
+	    Binary(state, OpBitwiseAnd, TypeU32(state), page, ConstantU32(state, 31)));
+	const auto pointer = FaultElementPointer(state, word);
+	const auto value   = state.builder.AllocateId();
+	state.builder.AddFunction({OpLoad, TypeU32(state), value, pointer});
+	state.builder.AddFunction(
+	    {OpStore, pointer, Binary(state, OpBitwiseOr, TypeU32(state), value, bit)});
+}
+
+uint32_t GetBdaPointer(ValueEmitContext& ctx, uint32_t address) {
+	auto&      state  = ctx.state;
+	const auto result = state.builder.AllocateId();
+	state.builder.AddFunction(
+	    {OpFunctionCall, TypeDeviceAddress(state), result, state.bda_pointer_function, address});
+	return result;
+}
+
+uint32_t LoadBdaDword(ValueEmitContext& ctx, uint32_t address) {
+	auto&      state   = ctx.state;
+	const auto bda     = GetBdaPointer(ctx, address);
+	const auto present = Binary(state, OpINotEqual, TypeBool(state), bda,
+	                            ConstantDeviceAddress(state, 0));
+	return EmitValueOrZeroIfCondition(state, present, [&]() {
+		const auto pointer = state.builder.AllocateId();
+		state.builder.AddFunction(
+		    {OpConvertUToPtr, TypePhysicalU32Pointer(state), pointer, bda});
+		const auto value = state.builder.AllocateId();
+		state.builder.AddFunction(
+		    {OpLoad, TypeU32(state), value, pointer, MemoryAccessAlignedMask, sizeof(uint32_t)});
+		return value;
+	});
+}
+
+uint32_t LoadBda(ValueEmitContext& ctx, const IR::Inst& inst, const IR::MemoryInfo& mem,
+	             uint32_t bits) {
+	auto&      state   = ctx.state;
+	const auto address = GuestAddress(ctx, inst, mem);
+	const auto active  = ctx.Arg(inst, inst.NumArgs() - 1);
+	return EmitValueOrZeroIfCondition(state, active, [&]() {
+		const auto aligned = Binary(state, OpBitwiseAnd, TypeDeviceAddress(state), address,
+		                            ConstantDeviceAddress(state, ~uint64_t {3}));
+		const auto first   = LoadBdaDword(ctx, aligned);
+		const auto byte = Binary(state, OpBitwiseAnd, TypeU32(state),
+		                         Unary(state, OpUConvert, TypeU32(state), address),
+		                         ConstantU32(state, 3));
+		const auto crosses = bits == 8u
+		                          ? ConstantBool(state, false)
+		                          : Binary(state, bits == 16u ? OpUGreaterThan : OpINotEqual,
+		                                   TypeBool(state), byte,
+		                                   ConstantU32(state, bits == 16u ? 2u : 0u));
+		const auto second = EmitValueOrZeroIfCondition(state, crosses, [&]() {
+			return LoadBdaDword(
+			    ctx, Binary(state, OpIAdd, TypeDeviceAddress(state), aligned,
+			                ConstantDeviceAddress(state, sizeof(uint32_t))));
+		});
+		const auto shift = Binary(state, OpShiftLeftLogical, TypeU32(state), byte,
+		                          ConstantU32(state, 3));
+		const auto upper_shift = Binary(
+		    state, OpShiftLeftLogical, TypeU32(state),
+		    Binary(state, OpBitwiseAnd, TypeU32(state),
+		           Binary(state, OpISub, TypeU32(state), ConstantU32(state, 4), byte),
+		           ConstantU32(state, 3)),
+		    ConstantU32(state, 3));
+		const auto merged = Binary(
+		    state, OpBitwiseOr, TypeU32(state),
+		    Binary(state, OpShiftRightLogical, TypeU32(state), first, shift),
+		    Binary(state, OpShiftLeftLogical, TypeU32(state), second, upper_shift));
+		return bits == 32u
+		           ? merged
+		           : Binary(state, OpBitwiseAnd, TypeU32(state), merged,
+		                    ConstantU32(state, bits == 8u ? 0xffu : 0xffffu));
+	});
 }
 
 uint32_t ByteAddress(ValueEmitContext& ctx, const IR::Inst& inst, const IR::MemoryInfo& mem) {
@@ -163,11 +264,10 @@ uint32_t ByteAddress(ValueEmitContext& ctx, const IR::Inst& inst, const IR::Memo
 		return Binary(ctx.state, OpIAdd, TypeU32(ctx.state), ctx.Arg(inst, 0),
 		              ConstantU32(ctx.state, mem.offset));
 	}
-	const auto address = AddressByteAddress(ctx, inst, mem, ctx.Arg(inst, 1), ctx.Arg(inst, 2));
-	if (mem.kind == IR::ResourceKind::Scratch) {
-		return address;
+	if (mem.kind != IR::ResourceKind::Scratch) {
+		EXIT("physical address memory must use the BDA emitter\n");
 	}
-	return AddAddressMemoryOffset(ctx, mem, address);
+	return ScratchByteAddress(ctx, mem, ctx.Arg(inst, 1), ctx.Arg(inst, 2));
 }
 
 uint32_t DwordIndex(ValueEmitContext& ctx, const IR::Inst& inst, const IR::MemoryInfo& mem) {
@@ -436,12 +536,12 @@ void StoreSubwordInBounds(ValueEmitContext& ctx, const IR::MemoryInfo& mem,
 		                     Unary(ctx.state, OpNot, TypeU32(ctx.state), mask)),
 		              value);
 	};
-	if (mem.kind == IR::ResourceKind::Lds || mem.kind == IR::ResourceKind::Gds) {
-		AtomicUpdate(ctx.state, pointer, mem.kind, merge);
-	} else {
+	if (mem.kind == IR::ResourceKind::Scratch) {
 		const auto old = ctx.state.builder.AllocateId();
 		ctx.state.builder.AddFunction({OpLoad, TypeU32(ctx.state), old, pointer});
 		ctx.state.builder.AddFunction({OpStore, pointer, merge(old)});
+	} else {
+		AtomicUpdate(ctx.state, pointer, mem.kind, merge);
 	}
 }
 
@@ -672,11 +772,9 @@ uint32_t AppendConsume(ValueEmitContext& ctx, const IR::Inst& inst, bool append)
 	const auto high = state.builder.AllocateId();
 	state.builder.AddFunction({OpCompositeExtract, TypeU32(state), low, ballot, 0});
 	state.builder.AddFunction({OpCompositeExtract, TypeU32(state), high, ballot, 1});
-	const auto count_low  = state.per_invocation_masks ? low : ctx.Arg(inst, 2);
-	const auto count_high = state.per_invocation_masks ? high : ctx.Arg(inst, 3);
 	const auto count =
-	    Binary(state, OpIAdd, TypeU32(state), Unary(state, OpBitCount, TypeU32(state), count_low),
-	           Unary(state, OpBitCount, TypeU32(state), count_high));
+	    Binary(state, OpIAdd, TypeU32(state), Unary(state, OpBitCount, TypeU32(state), low),
+	           Unary(state, OpBitCount, TypeU32(state), high));
 	const auto first = state.builder.AllocateId();
 	state.builder.AddFunction({OpGroupNonUniformBallotFindLSB, TypeU32(state), first,
 	                           ConstantU32(state, ScopeSubgroup), ballot});
@@ -925,6 +1023,58 @@ void StoreWideShared(ValueEmitContext& ctx, const IR::Inst& inst, uint32_t compo
 
 } // namespace
 
+void DefineGetBdaPointer(EmitterState& state) {
+	if (!state.program.info.uses_dma) {
+		return;
+	}
+	const auto type            = TypeDeviceAddress(state);
+	const auto function_type   = state.builder.Type(OpTypeFunction, {type, type});
+	state.bda_pointer_function = state.builder.AllocateId();
+	const auto address         = state.builder.AllocateId();
+	const auto entry_label     = state.builder.AllocateId();
+	state.builder.AddName(state.bda_pointer_function, "get_bda_pointer");
+	state.builder.AddFunction(
+	    {OpFunction, type, state.bda_pointer_function, FunctionControlNone, function_type});
+	state.builder.AddFunction({OpFunctionParameter, type, address});
+	EmitLabel(state, entry_label);
+
+	const auto page64 = Binary(
+	    state, OpShiftRightLogical, type, address,
+	    ConstantDeviceAddress(state, BufferCache::CACHING_PAGEBITS));
+	const auto page = Unary(state, OpUConvert, TypeU32(state), page64);
+	const auto entry_pointer = state.builder.AllocateId();
+	state.builder.AddFunction({OpAccessChain, TypeDeviceAddressStoragePointer(state), entry_pointer,
+	                           state.bda_pagetable_variable, ConstantU32(state, 0), page});
+	const auto base = state.builder.AllocateId();
+	state.builder.AddFunction({OpLoad, type, base, entry_pointer});
+	const auto missing =
+	    Binary(state, OpIEqual, TypeBool(state), base, ConstantDeviceAddress(state, 0));
+	const auto fault_label     = state.builder.AllocateId();
+	const auto available_label = state.builder.AllocateId();
+	const auto merge_label     = state.builder.AllocateId();
+	state.builder.AddFunction({OpSelectionMerge, merge_label, SelectionControlNone});
+	state.builder.AddFunction(
+	    {OpBranchConditional, missing, fault_label, available_label});
+
+	EmitLabel(state, fault_label);
+	RecordBdaFault(state, page);
+	state.builder.AddFunction({OpBranch, merge_label});
+
+	EmitLabel(state, available_label);
+	const auto offset = Binary(
+	    state, OpBitwiseAnd, type, address,
+	    ConstantDeviceAddress(state, BufferCache::CACHING_PAGESIZE - 1));
+	const auto available = Binary(state, OpIAdd, type, base, offset);
+	state.builder.AddFunction({OpBranch, merge_label});
+
+	EmitLabel(state, merge_label);
+	const auto result = state.builder.AllocateId();
+	state.builder.AddFunction({OpPhi, type, result, ConstantDeviceAddress(state, 0), fault_label,
+	                           available, available_label});
+	state.builder.AddFunction({OpReturnValue, result});
+	state.builder.AddFunction({OpFunctionEnd});
+}
+
 bool EmitValueMemory(ValueEmitContext& ctx, const IR::Inst& inst) {
 	auto&      state             = ctx.state;
 	const auto op                = inst.GetOpcode();
@@ -983,6 +1133,10 @@ bool EmitValueMemory(ValueEmitContext& ctx, const IR::Inst& inst) {
 	}
 	const auto address_info = IR::AddressOpcodeInfoOf(op);
 	const bool load_address = address_info.access == IR::AddressAccess::Read;
+	if (load_address && ctx.Memory(inst).kind != IR::ResourceKind::Scratch) {
+		ctx.Define(inst, LoadBda(ctx, inst, ctx.Memory(inst), address_info.data_bits));
+		return true;
+	}
 	const bool load_buffer  = op == IR::ValueOpcode::LoadBufferU8 ||
 	                          op == IR::ValueOpcode::LoadBufferU16 ||
 	                          op == IR::ValueOpcode::LoadBufferU32;
@@ -1040,24 +1194,27 @@ bool EmitValueMemory(ValueEmitContext& ctx, const IR::Inst& inst) {
 		ctx.Define(inst, AppendConsume(ctx, inst, op == IR::ValueOpcode::DataAppend));
 		return true;
 	}
-	if (op == IR::ValueOpcode::SwizzleU32) {
-		ctx.state.builder.AddFunction({OpStore, ctx.scratch_u32_variable, ctx.Arg(inst, 0)});
-		const auto source = ctx.state.builder.AllocateId();
-		ctx.state.builder.AddFunction(
-		    {OpLoad, TypeU32(ctx.state), source, ctx.scratch_u32_variable});
-		const auto target =
-		    EmitDsSwizzleTargetLane(state, EmitSubgroupLocalInvocationId(state),
-		                            inst.Arg(1).IsImmediate() ? inst.Arg(1).U32() : 0);
-		const auto shuffled = state.builder.AllocateId();
-		state.builder.AddFunction({OpGroupNonUniformShuffle, TypeU32(state), shuffled,
-		                           ConstantU32(state, ScopeSubgroup), source, target});
-		const auto source_exec = state.builder.AllocateId();
-		state.builder.AddFunction({OpGroupNonUniformShuffle, TypeBool(state), source_exec,
-		                           ConstantU32(state, ScopeSubgroup), ctx.Arg(inst, 2), target});
-		const auto source_active =
-		    AndCondition(state, source_exec, EmitSubgroupLaneActiveBool(state, target));
-		ctx.Define(inst,
-		           Select(state, TypeU32(state), source_active, shuffled, ConstantU32(state, 0)));
+	if (op == IR::ValueOpcode::SwizzleU32 || op == IR::ValueOpcode::BpermuteU32) {
+		uint32_t source = ctx.Arg(inst, 0);
+		uint32_t target = 0;
+		if (op == IR::ValueOpcode::SwizzleU32) {
+			state.builder.AddFunction({OpStore, ctx.scratch_u32_variable, source});
+			source = state.builder.AllocateId();
+			state.builder.AddFunction({OpLoad, TypeU32(state), source, ctx.scratch_u32_variable});
+			target = EmitDsSwizzleTargetLane(state, EmitSubgroupLocalInvocationId(state),
+			                                 inst.Arg(1).IsImmediate() ? inst.Arg(1).U32() : 0);
+		} else {
+			const auto index = Binary(
+			    state, OpBitwiseAnd, TypeU32(state),
+			    Binary(state, OpShiftRightLogical, TypeU32(state), ctx.Arg(inst, 1),
+			           ConstantU32(state, 2)),
+			    ConstantU32(state, 31));
+			const auto base =
+			    Binary(state, OpBitwiseAnd, TypeU32(state), EmitSubgroupLocalInvocationId(state),
+			           ConstantU32(state, ~31u));
+			target = Binary(state, OpBitwiseOr, TypeU32(state), base, index);
+		}
+		ctx.Define(inst, EmitDsMaskedLaneRead(state, source, target, ctx.Arg(inst, 2)));
 		return true;
 	}
 	return false;

src/graphics/shader/recompiler/backend/spirv/spirvEmitterMemoryHelpers.cpp

@@ -108,20 +108,8 @@ MemoryResourceAccess PrepareMemoryResourceAccess(EmitterState& state, const IR::
 			return access;
 		case IR::ResourceKind::ScalarAddress:
 		case IR::ResourceKind::Flat:
-		case IR::ResourceKind::Global: {
-			if (state.address_memory_variable == 0) {
-				ExitDescriptorBindingFailure(state, IR::DescriptorBindingKind::AddressMemory,
-				                             mem.resource,
-				                             "address memory binding was not emitted");
-			}
-			const auto binding = ResourceForDescriptor(
-			    state, IR::DescriptorBindingKind::AddressMemory, mem.resource);
-			access.object_pointer = state.builder.AllocateId();
-			state.builder.AddFunction({OpAccessChain, TypeStorageBufferPointer(state),
-			                           access.object_pointer, state.address_memory_variable,
-			                           ConstantU32(state, binding.array_index)});
-			break;
-		}
+		case IR::ResourceKind::Global:
+			EXIT("physical address memory must use the BDA emitter\n");
 		case IR::ResourceKind::ScalarBuffer:
 		case IR::ResourceKind::Buffer: {
 			if (state.storage_buffer_variable == 0) {

src/graphics/shader/recompiler/backend/spirv/spirvEmitterModule.cpp

@@ -28,6 +28,10 @@ uint32_t TypeU64(EmitterState& state) {
 	return TypeU32Vector(state, 2);
 }
 
+uint32_t TypeDeviceAddress(EmitterState& state) {
+	return state.builder.Type(OpTypeInt, {64, 0});
+}
+
 uint32_t TypeU32Pair(EmitterState& state) {
 	const auto element = TypeU32(state);
 	return state.builder.Type(OpTypeStruct, {element, element});
@@ -90,6 +94,14 @@ uint32_t TypeStorageBufferElementPointer(EmitterState& state) {
 	return TypePointer(state, StorageClassStorageBuffer, TypeU32(state));
 }
 
+uint32_t TypeDeviceAddressStoragePointer(EmitterState& state) {
+	return TypePointer(state, StorageClassStorageBuffer, TypeDeviceAddress(state));
+}
+
+uint32_t TypePhysicalU32Pointer(EmitterState& state) {
+	return TypePointer(state, StorageClassPhysicalStorageBuffer, TypeU32(state));
+}
+
 uint32_t TypePushConstantElementPointer(EmitterState& state) {
 	return TypePointer(state, StorageClassPushConstant, TypeU32(state));
 }
@@ -130,6 +142,19 @@ uint32_t SampleMaskArrayType(EmitterState& state) {
 	return state.builder.Type(OpTypeArray, {TypeI32(state), ConstantU32(state, 1)});
 }
 
+uint32_t BdaPagetableType(EmitterState& state) {
+	const auto array = state.builder.DecoratedType(
+	    OpTypeRuntimeArray, {TypeDeviceAddress(state)},
+	    {{OpDecorate, {DecorationArrayStride, sizeof(uint64_t)}}});
+	return state.builder.DecoratedType(
+	    OpTypeStruct, {array},
+	    {{OpMemberDecorate, {0, DecorationOffset, 0}}, {OpDecorate, {DecorationBlock}}});
+}
+
+uint32_t F32ArrayType(EmitterState& state, uint32_t count) {
+	return state.builder.Type(OpTypeArray, {TypeF32(state), ConstantU32(state, count)});
+}
+
 void DefineDescriptorVariables(EmitterState& state) {
 	if (DescriptorBinding(state, IR::DescriptorBindingKind::Buffers) != nullptr) {
 		const auto count =
@@ -139,13 +164,14 @@ void DefineDescriptorVariables(EmitterState& state) {
 		state.storage_buffer_variable =
 		    state.builder.DefineGlobalVariable(pointer_type, StorageClassStorageBuffer);
 	}
-	if (DescriptorBinding(state, IR::DescriptorBindingKind::AddressMemory) != nullptr) {
-		const auto count =
-		    ConstantU32(state, DescriptorCount(state, IR::DescriptorBindingKind::AddressMemory));
-		const auto array_type = state.builder.Type(OpTypeArray, {StorageBufferType(state), count});
-		const auto pointer_type = TypePointer(state, StorageClassStorageBuffer, array_type);
-		state.address_memory_variable =
-		    state.builder.DefineGlobalVariable(pointer_type, StorageClassStorageBuffer);
+	if (DescriptorBinding(state, IR::DescriptorBindingKind::BdaPagetable) != nullptr) {
+		state.bda_pagetable_variable = state.builder.DefineGlobalVariable(
+		    TypePointer(state, StorageClassStorageBuffer, BdaPagetableType(state)),
+		    StorageClassStorageBuffer);
+	}
+	if (DescriptorBinding(state, IR::DescriptorBindingKind::FaultBuffer) != nullptr) {
+		state.fault_buffer_variable = state.builder.DefineGlobalVariable(
+		    TypeStorageBufferPointer(state), StorageClassStorageBuffer);
 	}
 	if (state.program.bindings.push_constant_size != 0) {
 		const auto pointer_type =
@@ -420,6 +446,23 @@ void AllocateOutputVariables(EmitterState& state) {
 				binding.variable_id =
 				    AllocateSharedOutputVariable(state, state.per_vertex_variable);
 				break;
+			case IR::StageOutputKind::PointSize:
+				binding.variable_id =
+				    AllocateSharedOutputVariable(state, state.point_size_variable);
+				break;
+			case IR::StageOutputKind::ClipDistance:
+				binding.variable_id =
+				    AllocateSharedOutputVariable(state, state.clip_distance_variable);
+				state.clip_distance_count = std::max(state.clip_distance_count, binding.index + 1);
+				break;
+			case IR::StageOutputKind::CullDistance:
+				binding.variable_id =
+				    AllocateSharedOutputVariable(state, state.cull_distance_variable);
+				state.cull_distance_count = std::max(state.cull_distance_count, binding.index + 1);
+				break;
+			case IR::StageOutputKind::Layer:
+				binding.variable_id = AllocateSharedOutputVariable(state, state.layer_variable);
+				break;
 			case IR::StageOutputKind::Depth:
 				binding.variable_id = AllocateSharedOutputVariable(state, state.depth_variable);
 				break;
@@ -441,6 +484,8 @@ uint32_t BuiltInForInput(IR::StageInputKind kind) {
 		case IR::StageInputKind::InstanceIndex: return BuiltInInstanceIndex;
 		case IR::StageInputKind::FragCoord: return BuiltInFragCoord;
 		case IR::StageInputKind::FrontFacing: return BuiltInFrontFacing;
+		case IR::StageInputKind::BaryCoordSmooth: return BuiltInBaryCoordKHR;
+		case IR::StageInputKind::BaryCoordNoPerspective: return BuiltInBaryCoordNoPerspKHR;
 		case IR::StageInputKind::WorkgroupId: return BuiltInWorkgroupId;
 		case IR::StageInputKind::LocalInvocationId: return BuiltInLocalInvocationId;
 		case IR::StageInputKind::LocalInvocationIndex: return BuiltInLocalInvocationIndex;
@@ -464,11 +509,14 @@ void AddInputAnnotationsAndNames(EmitterState& state) {
 		state.builder.AddName(input.variable_id, input.debug_name.c_str());
 		if (input.kind == IR::StageInputKind::Parameter) {
 			const auto flat = PixelParameterIsFlat(state, input.location);
-			if (flat) {
+			if (input.per_vertex) {
+				state.builder.AddAnnotation(
+				    {OpDecorate, input.variable_id, DecorationPerVertexKHR});
+			} else if (flat) {
 				state.builder.AddAnnotation({OpDecorate, input.variable_id, DecorationFlat});
 			}
 			if (state.stage == ShaderType::Pixel && state.input_info.pixel->ps_no_perspective &&
-			    !flat) {
+			    !flat && !input.per_vertex) {
 				state.builder.AddAnnotation(
 				    {OpDecorate, input.variable_id, DecorationNoPerspective});
 			}
@@ -490,6 +538,16 @@ void AddOutputAnnotationsAndNames(EmitterState& state) {
 		state.builder.AddName(PerVertexType(state), "gl_PerVertex");
 		state.builder.AddName(state.per_vertex_variable, "outPerVertex");
 	}
+	auto AddBuiltIn = [&](uint32_t variable, const char* name, uint32_t builtin) {
+		if (variable != 0) {
+			state.builder.AddName(variable, name);
+			state.builder.AddAnnotation({OpDecorate, variable, DecorationBuiltIn, builtin});
+		}
+	};
+	AddBuiltIn(state.point_size_variable, "gl_PointSize", BuiltInPointSize);
+	AddBuiltIn(state.clip_distance_variable, "gl_ClipDistance", BuiltInClipDistance);
+	AddBuiltIn(state.cull_distance_variable, "gl_CullDistance", BuiltInCullDistance);
+	AddBuiltIn(state.layer_variable, "gl_Layer", BuiltInLayer);
 	if (state.depth_variable != 0) {
 		state.builder.AddName(state.depth_variable, "gl_FragDepth");
 		state.builder.AddAnnotation(
@@ -510,30 +568,31 @@ void AddOutputAnnotationsAndNames(EmitterState& state) {
 	}
 }
 
-void DecorateDescriptor(EmitterState& state, uint32_t variable, const char* name, uint32_t set,
-                        uint32_t binding) {
+void DecorateDescriptor(EmitterState& state, uint32_t variable, const char* name,
+                        IR::DescriptorBindingKind kind) {
 	if (variable == 0) {
 		return;
 	}
 	state.builder.AddName(variable, name);
-	state.builder.AddAnnotation({OpDecorate, variable, DecorationDescriptorSet, set});
-	state.builder.AddAnnotation({OpDecorate, variable, DecorationBinding, binding});
+	state.builder.AddAnnotation({OpDecorate, variable, DecorationDescriptorSet, 0});
+	state.builder.AddAnnotation(
+	    {OpDecorate, variable, DecorationBinding, IR::NativeBinding(state.program.stage, kind)});
 }
 
 void AddDescriptorAnnotationsAndNames(EmitterState& state) {
-	const auto set      = state.program.bindings.descriptor_set;
-	auto       Decorate = [&](uint32_t variable, const char* name, IR::DescriptorBindingKind kind) {
-		const auto* binding = DescriptorBinding(state, kind);
-		if (binding != nullptr) {
-			DecorateDescriptor(state, variable, name, set, binding->binding);
-		}
+	auto Decorate = [&](uint32_t variable, const char* name, IR::DescriptorBindingKind kind) {
+		DecorateDescriptor(state, variable, name, kind);
 	};
 	if (state.storage_buffer_variable != 0) {
 		Decorate(state.storage_buffer_variable, "buffers", IR::DescriptorBindingKind::Buffers);
 	}
-	if (state.address_memory_variable != 0) {
-		Decorate(state.address_memory_variable, "address_memory",
-		         IR::DescriptorBindingKind::AddressMemory);
+	if (state.bda_pagetable_variable != 0) {
+		Decorate(state.bda_pagetable_variable, "bda_pagetable",
+		         IR::DescriptorBindingKind::BdaPagetable);
+	}
+	if (state.fault_buffer_variable != 0) {
+		Decorate(state.fault_buffer_variable, "fault_buffer",
+		         IR::DescriptorBindingKind::FaultBuffer);
 	}
 	constexpr const char* SampledNames[] = {"sampled_1d",
 	                                        "sampled_1d_array",
@@ -570,7 +629,7 @@ void AddDescriptorAnnotationsAndNames(EmitterState& state) {
 	                                        "storage_atomic_2d_array",
 	                                        "storage_atomic_3d"};
 	for (uint32_t i = 0; i < state.storage_image_variables.size(); i++) {
-		const auto view = static_cast<ImageViewKind>(i % StorageImageViewKindCount);
+		const auto view        = static_cast<ImageViewKind>(i % StorageImageViewKindCount);
 		const auto image_class = static_cast<StorageImageClass>(i / StorageImageViewKindCount);
 		Decorate(state.storage_image_variables[i], StorageNames[i],
 		         StorageBindingKind(image_class, view));
@@ -588,17 +647,13 @@ void AddDescriptorAnnotationsAndNames(EmitterState& state) {
 }
 
 void AddVsharpAnnotationsAndNames(EmitterState& state) {
-	const auto& bind = state.program.bindings;
 	if (state.push_constant_variable != 0) {
 		state.builder.AddName(PushConstantBlockType(state), "BufferResource");
 		state.builder.AddName(state.push_constant_variable, "vsharp");
 	}
 	if (state.vsharp_storage_variable != 0) {
-		const auto* storage = DescriptorBinding(state, IR::DescriptorBindingKind::UserData);
-		if (storage != nullptr) {
-			DecorateDescriptor(state, state.vsharp_storage_variable, "user_data",
-			                   bind.descriptor_set, storage->binding);
-		}
+		DecorateDescriptor(state, state.vsharp_storage_variable, "user_data",
+		                   IR::DescriptorBindingKind::UserData);
 	}
 }
 
@@ -615,6 +670,21 @@ void DefineModule(EmitterState& state) {
 
 	state.builder.RequireCapability(CapabilityShader);
 	state.builder.RequireCapability(CapabilitySignedZeroInfNanPreserve);
+	if (state.program.info.uses_dma) {
+		state.builder.RequireCapability(CapabilityInt64);
+		state.builder.RequireCapability(CapabilityPhysicalStorageBufferAddresses);
+		state.builder.RequireExtension("SPV_KHR_physical_storage_buffer");
+	}
+	if (state.clip_distance_variable != 0) {
+		state.builder.RequireCapability(CapabilityClipDistance);
+	}
+	if (state.cull_distance_variable != 0) {
+		state.builder.RequireCapability(CapabilityCullDistance);
+	}
+	if (state.layer_variable != 0) {
+		state.builder.RequireCapability(CapabilityShaderViewportIndexLayerEXT);
+		state.builder.RequireExtension("SPV_EXT_shader_viewport_index_layer");
+	}
 	if (state.requirements.image_gather_extended) {
 		state.builder.RequireCapability(CapabilityImageGatherExtended);
 	}
@@ -632,8 +702,21 @@ void DefineModule(EmitterState& state) {
 		state.builder.RequireCapability(CapabilityComputeDerivativeGroupQuadsKHR);
 		state.builder.RequireExtension("SPV_KHR_compute_shader_derivatives");
 	}
+	const bool fragment_barycentric =
+	    state.stage == ShaderType::Pixel &&
+	    std::any_of(state.inputs.begin(), state.inputs.end(), [](const InputBinding& input) {
+		    return input.per_vertex || input.kind == IR::StageInputKind::BaryCoordSmooth ||
+		           input.kind == IR::StageInputKind::BaryCoordNoPerspective;
+	    });
+	if (fragment_barycentric) {
+		state.builder.RequireCapability(CapabilityFragmentBarycentricKHR);
+		state.builder.RequireExtension("SPV_KHR_fragment_shader_barycentric");
+	}
 	state.builder.RequireExtension("SPV_KHR_float_controls");
-	state.builder.AddMemoryModel({AddressingModelLogical, MemoryModelGLSL450});
+	state.builder.AddMemoryModel(
+	    {state.program.info.uses_dma ? AddressingModelPhysicalStorageBuffer64
+	                                 : AddressingModelLogical,
+	     MemoryModelGLSL450});
 	state.builder.AddEntryPoint(ExecutionModelForStage(state.stage), state.main_func, "main",
 	                            state.interface_variables);
 	// GCN/RDNA arithmetic preserves 32-bit signed zero, infinity, and NaN. Declaring that
@@ -696,6 +779,10 @@ void DefineModule(EmitterState& state) {
 			case IR::StageInputKind::FragCoord:
 				ptr_type = TypePointer(state, StorageClassInput, TypeF32Vector(state, 4));
 				break;
+			case IR::StageInputKind::BaryCoordSmooth:
+			case IR::StageInputKind::BaryCoordNoPerspective:
+				ptr_type = TypePointer(state, StorageClassInput, TypeF32Vector(state, 3));
+				break;
 			case IR::StageInputKind::FrontFacing:
 				ptr_type = TypePointer(state, StorageClassInput, TypeBool(state));
 				break;
@@ -704,6 +791,10 @@ void DefineModule(EmitterState& state) {
 					const auto kind       = VertexParameterScalarKind(state, input.location);
 					const auto components = VertexParameterComponentCount(state, input);
 					ptr_type = VertexParameterInputPointerType(state, kind, components);
+				} else if (input.per_vertex) {
+					const auto array_type = state.builder.Type(
+					    OpTypeArray, {TypeF32Vector(state, 4), ConstantU32(state, 3)});
+					ptr_type = TypePointer(state, StorageClassInput, array_type);
 				} else {
 					ptr_type = TypePointer(state, StorageClassInput, TypeF32Vector(state, 4));
 				}
@@ -717,6 +808,30 @@ void DefineModule(EmitterState& state) {
 		    state.per_vertex_variable, TypePointer(state, StorageClassOutput, PerVertexType(state)),
 		    StorageClassOutput);
 	}
+	if (state.point_size_variable != 0) {
+		state.builder.DefineGlobalVariable(
+		    state.point_size_variable, TypePointer(state, StorageClassOutput, TypeF32(state)),
+		    StorageClassOutput);
+	}
+	if (state.clip_distance_variable != 0) {
+		state.builder.DefineGlobalVariable(
+		    state.clip_distance_variable,
+		    TypePointer(state, StorageClassOutput,
+		                F32ArrayType(state, state.clip_distance_count)),
+		    StorageClassOutput);
+	}
+	if (state.cull_distance_variable != 0) {
+		state.builder.DefineGlobalVariable(
+		    state.cull_distance_variable,
+		    TypePointer(state, StorageClassOutput,
+		                F32ArrayType(state, state.cull_distance_count)),
+		    StorageClassOutput);
+	}
+	if (state.layer_variable != 0) {
+		state.builder.DefineGlobalVariable(
+		    state.layer_variable, TypePointer(state, StorageClassOutput, TypeU32(state)),
+		    StorageClassOutput);
+	}
 	for (const auto& binding: state.outputs) {
 		if (binding.kind == IR::StageOutputKind::Parameter ||
 		    binding.kind == IR::StageOutputKind::Mrt) {

src/graphics/shader/recompiler/backend/spirv/spirvEmitterProgram.cpp

@@ -1,8 +1,9 @@
 #include "graphics/shader/recompiler/backend/spirv/spirvEmitterInternal.h"
 
+#include "common/assert.h"
+
 #include <algorithm>
 #include <bit>
-#include <fmt/format.h>
 #include <functional>
 #include <optional>
 #include <unordered_set>
@@ -79,9 +80,9 @@ void StoreDispatcherPhiEdge(ValueEmitContext& ctx, const DispatcherFunctionState
 	}
 }
 
-const IR::Block* TargetBlock(const IR::ValueProgram& program, uint32_t id) {
+const IR::Block* TargetBlock(const IR::Program& program, uint32_t id) {
 	const auto found = std::ranges::find_if(
-	    program.block_info, [&](const IR::ValueBlockInfo& info) { return info.id == id; });
+	    program.block_info, [&](const IR::BlockInfo& info) { return info.id == id; });
 	if (found == program.block_info.end()) {
 		return nullptr;
 	}
@@ -94,7 +95,7 @@ void EmitReturn(ValueEmitContext& ctx) {
 }
 
 void EmitStructuredTerminator(ValueEmitContext& ctx, const IR::Block* block,
-                              const IR::ValueBlockInfo& info) {
+                              const IR::BlockInfo& info) {
 	const auto& term       = info.terminator;
 	const auto  emit_merge = [&]() {
 		if (term.loop_header) {
@@ -150,7 +151,7 @@ void EmitDispatcherTarget(ValueEmitContext& ctx, const DispatcherFunctionState&
 }
 
 uint32_t EmitDispatcherNextPc(ValueEmitContext& ctx, const DispatcherFunctionState& dispatcher,
-                              const IR::Block* block, const IR::ValueBlockInfo& info) {
+                              const IR::Block* block, const IR::BlockInfo& info) {
 	const auto& term = info.terminator;
 	switch (term.kind) {
 		case CFG::TerminatorKind::Branch:
@@ -196,57 +197,50 @@ uint32_t EmitDispatcherNextPc(ValueEmitContext& ctx, const DispatcherFunctionSta
 	}
 }
 
-bool EmitDirectInstruction(ValueEmitContext& ctx, const IR::Inst& inst) {
+void EmitDirectInstruction(ValueEmitContext& ctx, const IR::Inst& inst) {
 	if (EmitValueFlow(ctx, inst) || EmitValueAlu(ctx, inst) || EmitValueMemory(ctx, inst) ||
 	    EmitValueImage(ctx, inst)) {
-		return true;
+		return;
 	}
 	ctx.Fail(inst, "has no direct SPIR-V emitter");
-	return false;
 }
 
-bool EmitStructuredInstruction(ValueEmitContext& ctx, StructuredFunctionState& structured,
+void EmitStructuredInstruction(ValueEmitContext& ctx, StructuredFunctionState& structured,
                                const IR::Inst& inst) {
 	if (inst.GetOpcode() == IR::ValueOpcode::Phi) {
 		const auto type = ctx.TypeId(inst.GetType());
 		if (type == 0 || inst.NumArgs() == 0) {
 			ctx.Fail(inst, "has no native SPIR-V representation");
-			return false;
 		}
 		for (size_t index = 0; index < inst.NumArgs(); index++) {
 			const auto* predecessor = inst.PhiBlock(index);
 			if (predecessor == nullptr || !ctx.labels.contains(predecessor)) {
 				ctx.Fail(inst, "has a predecessor outside the structured function");
-				return false;
 			}
 		}
 		structured.deferred_phis.push_back(
 		    {ctx.state.builder.AddDeferredPhi(type, ctx.Result(inst), inst.NumArgs()), &inst});
-		return true;
+		return;
 	}
-	return EmitDirectInstruction(ctx, inst);
+	EmitDirectInstruction(ctx, inst);
 }
 
-bool EmitDispatcherInstruction(ValueEmitContext& ctx, const DispatcherFunctionState& dispatcher,
+void EmitDispatcherInstruction(ValueEmitContext& ctx, const DispatcherFunctionState& dispatcher,
                                const IR::Inst& inst) {
 	if (inst.GetOpcode() == IR::ValueOpcode::Phi) {
 		const auto type = ctx.TypeId(inst.GetType());
 		if (type == 0) {
 			ctx.Fail(inst, "cannot be loaded by the dispatcher");
-			return false;
 		}
 		ctx.state.builder.AddFunction(
 		    {OpLoad, type, ctx.Result(inst), dispatcher.spills.at(&inst)});
-		return true;
-	}
-	if (!EmitDirectInstruction(ctx, inst)) {
-		return false;
+		return;
 	}
+	EmitDirectInstruction(ctx, inst);
 	if (const auto found = dispatcher.spills.find(&inst); found != dispatcher.spills.end()) {
 		ctx.state.builder.AddFunction(
 		    {OpStore, found->second, ctx.Def(IR::Value(const_cast<IR::Inst*>(&inst)))});
 	}
-	return true;
 }
 
 template <typename EmitInstruction>
@@ -258,14 +252,11 @@ void EmitBlock(ValueEmitContext& ctx, const IR::Block* block, EmitInstruction&&
 		if (inst.GetOpcode() == IR::ValueOpcode::Phi) {
 			if (emitted_non_phi) {
 				ctx.Fail(inst, "appears after a non-Phi instruction");
-				return;
 			}
 		} else {
 			emitted_non_phi = true;
 		}
-		if (!emit_instruction(inst)) {
-			return;
-		}
+		emit_instruction(inst);
 	}
 }
 
@@ -275,15 +266,10 @@ void PatchStructuredPhis(ValueEmitContext& ctx, StructuredFunctionState& structu
 			const auto* predecessor = deferred.instruction->PhiBlock(index);
 			const auto  found       = structured.block_exit_labels.find(predecessor);
 			if (found == structured.block_exit_labels.end()) {
-				ctx.failed = true;
-				ctx.error  = "typed Phi predecessor was not emitted";
-				return;
+				ctx.Fail(*deferred.instruction, "has a predecessor that was not emitted");
 			}
 			ctx.state.builder.PatchDeferredPhi(
 			    deferred.phi, index, ctx.Def(deferred.instruction->Arg(index)), found->second);
-			if (ctx.failed) {
-				return;
-			}
 		}
 	}
 }
@@ -291,20 +277,15 @@ void PatchStructuredPhis(ValueEmitContext& ctx, StructuredFunctionState& structu
 void EmitStructuredFunction(ValueEmitContext& ctx) {
 	StructuredFunctionState structured;
 	ctx.state.builder.AddFunction({OpBranch, ctx.Label(ctx.program.blocks.front())});
-	for (size_t index = 0; index < ctx.program.blocks.size() && !ctx.failed; index++) {
+	for (size_t index = 0; index < ctx.program.blocks.size(); index++) {
 		const auto* block = ctx.program.blocks[index];
 		EmitBlock(ctx, block, [&](const IR::Inst& inst) {
-			return EmitStructuredInstruction(ctx, structured, inst);
+			EmitStructuredInstruction(ctx, structured, inst);
 		});
-		if (ctx.failed) {
-			break;
-		}
 		structured.block_exit_labels.emplace(block, ctx.state.current_label);
 		EmitStructuredTerminator(ctx, block, ctx.program.block_info[index]);
 	}
-	if (!ctx.failed) {
-		PatchStructuredPhis(ctx, structured);
-	}
+	PatchStructuredPhis(ctx, structured);
 }
 
 void EmitDispatcherFunction(ValueEmitContext& ctx, const DispatcherFunctionState& dispatcher) {
@@ -312,11 +293,8 @@ void EmitDispatcherFunction(ValueEmitContext& ctx, const DispatcherFunctionState
 	const auto* entry = ctx.program.blocks.front();
 	state.builder.AddFunction({OpBranch, ctx.Label(entry)});
 	EmitBlock(ctx, entry, [&](const IR::Inst& inst) {
-		return EmitDispatcherInstruction(ctx, dispatcher, inst);
+		EmitDispatcherInstruction(ctx, dispatcher, inst);
 	});
-	if (ctx.failed) {
-		return;
-	}
 	const auto initial_pc =
 	    EmitDispatcherNextPc(ctx, dispatcher, entry, ctx.program.block_info.front());
 	const auto initial_parent = state.current_label;
@@ -347,22 +325,16 @@ void EmitDispatcherFunction(ValueEmitContext& ctx, const DispatcherFunctionState
 	std::vector<uint32_t> next_pc_words {OpPhi, TypeU32(state), next_pc,
 	                                     ConstantU32(state, UINT32_MAX), dispatcher.select_label};
 
-	for (size_t index = 1; index < ctx.program.blocks.size() && !ctx.failed; index++) {
+	for (size_t index = 1; index < ctx.program.blocks.size(); index++) {
 		EmitBlock(ctx, ctx.program.blocks[index], [&](const IR::Inst& inst) {
-			return EmitDispatcherInstruction(ctx, dispatcher, inst);
+			EmitDispatcherInstruction(ctx, dispatcher, inst);
 		});
-		if (ctx.failed) {
-			break;
-		}
 		const auto selected = EmitDispatcherNextPc(ctx, dispatcher, ctx.program.blocks[index],
 		                                           ctx.program.block_info[index]);
 		next_pc_words.push_back(selected);
 		next_pc_words.push_back(state.current_label);
 		state.builder.AddFunction({OpBranch, dispatcher.after_switch_label});
 	}
-	if (ctx.failed) {
-		return;
-	}
 	EmitLabel(state, dispatcher.after_switch_label);
 	state.builder.AddFunction(next_pc_words);
 	state.builder.AddFunction({OpBranch, dispatcher.continue_label});
@@ -408,9 +380,7 @@ uint32_t ValueEmitContext::Def(IR::Value value) {
 	}
 	const auto* inst = value.ResolveInstruction();
 	if (inst == nullptr) {
-		failed = true;
-		error  = "direct SPIR-V emitter received a non-value argument";
-		return ConstantU32(state, 0);
+		Fail("direct SPIR-V emitter received a non-value argument");
 	}
 	if (dispatcher_spills != nullptr && current_block != nullptr &&
 	    inst->Parent() != current_block) {
@@ -469,9 +439,7 @@ uint32_t ValueEmitContext::Define(const IR::Inst& inst, uint32_t value) {
 uint32_t ValueEmitContext::ResourceIndex(IR::Value value, IR::ValueOpcode opcode) {
 	const auto* inst = value.ResolveInstruction();
 	if (inst == nullptr || inst->GetOpcode() != opcode) {
-		failed = true;
-		error  = "typed resource handle has the wrong producer";
-		return 0;
+		Fail("typed resource handle has the wrong producer");
 	}
 	return inst->Flags<uint32_t>();
 }
@@ -479,9 +447,7 @@ uint32_t ValueEmitContext::ResourceIndex(IR::Value value, IR::ValueOpcode opcode
 const IR::Inst* ValueEmitContext::ImageAddress(IR::Value value) {
 	const auto* inst = value.ResolveInstruction();
 	if (inst == nullptr || inst->GetOpcode() != IR::ValueOpcode::MakeImageAddress) {
-		failed = true;
-		error  = "typed image address was not constructed by MakeImageAddress";
-		return nullptr;
+		Fail("typed image address was not constructed by MakeImageAddress");
 	}
 	return inst;
 }
@@ -498,12 +464,20 @@ uint32_t ValueEmitContext::Label(const IR::Block* block) const {
 	return labels.at(block);
 }
 
-void ValueEmitContext::Fail(const IR::Inst& inst, const char* reason) {
-	failed = true;
-	error  = fmt::format("typed opcode {} {}", IR::ValueOpcodeName(inst.GetOpcode()), reason);
+[[noreturn]] void ValueEmitContext::Fail(const char* reason) const {
+	EXIT("SPIR-V emission failed: hash=0x%016" PRIx64 " stage=%u reason=%s\n",
+	     program.shader_hash, static_cast<unsigned>(program.stage), reason);
+	std::abort();
 }
 
-bool EmitValueProgram(EmitterState& state, const IR::ValueProgram& program, std::string* error) {
+[[noreturn]] void ValueEmitContext::Fail(const IR::Inst& inst, const char* reason) const {
+	EXIT("SPIR-V emission failed: hash=0x%016" PRIx64 " stage=%u opcode=%s reason=%s\n",
+	     program.shader_hash, static_cast<unsigned>(program.stage),
+	     IR::ValueOpcodeName(inst.GetOpcode()), reason);
+	std::abort();
+}
+
+void EmitProgram(EmitterState& state, const IR::Program& program) {
 	ValueEmitContext                       ctx(state, program);
 	std::optional<DispatcherFunctionState> dispatcher;
 	if (state.stage == ShaderType::Pixel && state.requirements.pixel_valid_mask) {
@@ -562,10 +536,7 @@ bool EmitValueProgram(EmitterState& state, const IR::ValueProgram& program, std:
 		dispatch.merge_label        = state.builder.AllocateId();
 		ctx.dispatcher_spills       = &dispatch.spills;
 	}
-	if (ctx.failed) {
-		SetError(error, ctx.error.c_str());
-		return false;
-	}
+	DefineGetBdaPointer(state);
 	for (const auto* block: program.blocks) {
 		if (std::ranges::any_of(*block, [](const IR::Inst& inst) {
 			    return inst.GetOpcode() == IR::ValueOpcode::SwizzleU32 ||
@@ -629,11 +600,6 @@ bool EmitValueProgram(EmitterState& state, const IR::ValueProgram& program, std:
 		EmitStructuredFunction(ctx);
 	}
 	state.builder.AddFunction({OpFunctionEnd});
-	if (ctx.failed) {
-		SetError(error, ctx.error.c_str());
-		return false;
-	}
-	return true;
 }
 
 } // namespace Libs::Graphics::ShaderRecompiler::Spirv::Emitter

src/graphics/shader/recompiler/frontend/cfg/ShaderCFG.cpp

@@ -1,5 +1,7 @@
 #include "graphics/shader/recompiler/frontend/cfg/ShaderCFG.h"
 
+#include "common/assert.h"
+
 #include <algorithm>
 #include <fmt/format.h>
 #include <iterator>
@@ -24,19 +26,18 @@ struct SetpcTargetInfo {
 	std::vector<uint32_t> selector_target_pcs;
 };
 
-void SetError(std::string* error, const std::string& message) {
-	if (error != nullptr) {
-		*error = message;
-	}
-}
-
-void SetFailure(Graph& graph, FailureKind kind, uint32_t block_id, const std::string& message,
-                std::string* error) {
+void SetFailure(Graph& graph, FailureKind kind, uint32_t block_id, const std::string& message) {
 	graph.unsupported        = true;
 	graph.failure_kind       = kind;
 	graph.failure_block      = block_id;
 	graph.unsupported_reason = message;
-	SetError(error, message);
+}
+
+[[noreturn]] void ExitBuildFailure(Graph& graph, FailureKind kind, uint32_t block_id,
+                                   const std::string& message) {
+	SetFailure(graph, kind, block_id, message);
+	EXIT("shader CFG build failed: %s", message.c_str());
+	std::abort();
 }
 
 uint32_t InstructionEndPc(const Instruction& inst) {
@@ -1267,7 +1268,7 @@ bool MergeLeavesContainingLoop(const Graph& graph, uint32_t header, uint32_t mer
 	return false;
 }
 
-bool CanonicalizeNaturalLoops(Graph& graph, std::string* error) {
+bool CanonicalizeNaturalLoops(Graph& graph) {
 	const auto rewrite_budget = graph.blocks.size() * 2u + 16u;
 	for (size_t rewrite = 0; rewrite < rewrite_budget; rewrite++) {
 		bool changed = false;
@@ -1322,19 +1323,18 @@ bool CanonicalizeNaturalLoops(Graph& graph, std::string* error) {
 	}
 
 	SetFailure(graph, FailureKind::StructuredControlFlow, graph.entry_block,
-	           "CFG loop canonicalization exceeded rewrite budget", error);
+	           "CFG loop canonicalization exceeded rewrite budget");
 	return false;
 }
 
-bool IsolateSemanticLoopHeaders(Graph& graph, std::string* error) {
+bool IsolateSemanticLoopHeaders(Graph& graph) {
 	const auto isolation_budget = graph.natural_loops.size() + 1u;
 	for (size_t isolation = 0; isolation < isolation_budget; isolation++) {
-		const auto loop = std::find_if(graph.natural_loops.begin(), graph.natural_loops.end(),
-		                               [&](const auto& value) {
-			                               const auto* header = graph.FindBlock(value.header);
-			                               return header != nullptr &&
-			                                      header->inst_begin != header->inst_end;
-		                               });
+		const auto loop = std::find_if(
+		    graph.natural_loops.begin(), graph.natural_loops.end(), [&](const auto& value) {
+			    const auto* header = graph.FindBlock(value.header);
+			    return header != nullptr && header->inst_begin != header->inst_end;
+		    });
 		if (loop == graph.natural_loops.end()) {
 			return true;
 		}
@@ -1342,18 +1342,17 @@ bool IsolateSemanticLoopHeaders(Graph& graph, std::string* error) {
 		// SPIR-V requires OpLoopMerge and its branch to remain in the loop header's
 		// physical block. Keep that header as a dedicated control node, exactly like
 		// the separate Loop node in shadPS4's structured AST. Guest instructions live
-		// in the body so later lowering may introduce bounds/EXEC control flow without
+		// in the body so later translation may introduce bounds/EXEC control flow without
 		// displacing OpLoopMerge into a helper-created block.
 		if (!IsolateSemanticLoopHeader(graph, loop->header)) {
 			SetFailure(graph, FailureKind::StructuredControlFlow, loop->header,
-			           fmt::format("failed to isolate semantic loop header {}", loop->header),
-			           error);
+			           fmt::format("failed to isolate semantic loop header {}", loop->header));
 			return false;
 		}
 	}
 
 	SetFailure(graph, FailureKind::StructuredControlFlow, graph.entry_block,
-	           "CFG semantic loop-header isolation exceeded rewrite budget", error);
+	           "CFG semantic loop-header isolation exceeded rewrite budget");
 	return false;
 }
 
@@ -1453,7 +1452,7 @@ std::vector<uint32_t> SelectionRegion(const Graph& graph, const BasicBlock& head
 	return region;
 }
 
-bool SplitOneSelectionMerge(Graph& graph, std::string* error) {
+bool SplitOneSelectionMerge(Graph& graph) {
 	std::vector<uint32_t> loop_headers;
 	loop_headers.reserve(graph.natural_loops.size());
 	for (const auto& loop: graph.natural_loops) {
@@ -1501,8 +1500,7 @@ bool SplitOneSelectionMerge(Graph& graph, std::string* error) {
 			    graph, FailureKind::StructuredControlFlow, block_id,
 			    fmt::format("selection header block {} has externally entered region block {}; "
 			                "semantic block cloning is disabled",
-			                block_id, *external),
-			    error);
+			                block_id, *external));
 			return false;
 		}
 		const auto construct_blocks = DominatedBlocks(graph, block_id, merge);
@@ -1514,11 +1512,11 @@ bool SplitOneSelectionMerge(Graph& graph, std::string* error) {
 	return false;
 }
 
-bool SplitSharedMergeBlocks(Graph& graph, std::string* error) {
+bool SplitSharedMergeBlocks(Graph& graph) {
 	const auto original_block_count = static_cast<uint32_t>(graph.blocks.size());
 	const auto split_budget         = std::max<uint32_t>(16u, original_block_count * 4u);
 	for (uint32_t splits = 0; splits < split_budget; splits++) {
-		if (!SplitOneLoopMerge(graph) && !SplitOneSelectionMerge(graph, error)) {
+		if (!SplitOneLoopMerge(graph) && !SplitOneSelectionMerge(graph)) {
 			return !graph.unsupported;
 		}
 		RebuildPredecessors(graph);
@@ -1528,8 +1526,7 @@ bool SplitSharedMergeBlocks(Graph& graph, std::string* error) {
 	           fmt::format("CFG shared merge splitting exceeded budget: original_blocks={} "
 	                       "current_blocks={} split_budget={}",
 	                       original_block_count, static_cast<uint64_t>(graph.blocks.size()),
-	                       split_budget),
-	           error);
+	                       split_budget));
 	return false;
 }
 
@@ -1875,7 +1872,8 @@ bool RouteSharedSelectionArm(Graph& graph, uint32_t route_variable) {
 		return false;
 	}
 	const auto block_count = static_cast<uint32_t>(graph.blocks.size());
-	for (uint32_t block_id = 0; block_id < block_count; block_id++) {
+	// Route later/deeper candidates first so earlier shared arms remain structured.
+	for (uint32_t block_id = block_count; block_id-- > 0;) {
 		GotoRouteBlocks route;
 		if (RouteOneSharedArm(graph, block_count, block_id, route_variable, route)) {
 			PlaceGotoRouteBlocks(graph, block_count, route);
@@ -1889,12 +1887,11 @@ bool RouteSharedSelectionArm(Graph& graph, uint32_t route_variable) {
 
 } // namespace
 
-bool BuildGraph(const Decoder::Program& program, Graph& graph, std::string* error) {
-	graph = {};
-
+Graph BuildGraph(const Decoder::Program& program) {
+	Graph graph;
 	if (program.instructions.empty()) {
-		SetError(error, "cannot build CFG for empty shader");
-		return false;
+		ExitBuildFailure(graph, FailureKind::InvalidLabel, UINT32_MAX,
+		                 "cannot build CFG for empty shader");
 	}
 
 	const auto first_pc = program.instructions.front().pc;
@@ -1904,11 +1901,10 @@ bool BuildGraph(const Decoder::Program& program, Graph& graph, std::string* erro
 	for (const auto& inst: program.instructions) {
 		instruction_pcs.insert(inst.pc);
 		if (inst.opcode == Opcode::UNSUPPORTED) {
-			SetFailure(graph, FailureKind::UnsupportedInstruction, UINT32_MAX,
-			           fmt::format("unsupported decoded instruction in CFG at pc 0x{:08x}: {}",
-			                       inst.pc, Decoder::InstructionToString(inst).c_str()),
-			           error);
-			return false;
+			ExitBuildFailure(
+			    graph, FailureKind::UnsupportedInstruction, UINT32_MAX,
+			    fmt::format("unsupported decoded instruction in CFG at pc 0x{:08x}: {}", inst.pc,
+			                Decoder::InstructionToString(inst).c_str()));
 		}
 	}
 
@@ -1922,11 +1918,9 @@ bool BuildGraph(const Decoder::Program& program, Graph& graph, std::string* erro
 		const auto  next_pc = InstructionEndPc(inst);
 		if (IsBranch(inst.opcode)) {
 			if (!IsValidTarget(inst.branch_target, instruction_pcs, first_pc, end_pc)) {
-				SetFailure(graph, FailureKind::InvalidBranchTarget, UINT32_MAX,
-				           fmt::format("branch at pc 0x{:08x} targets invalid pc 0x{:08x}", inst.pc,
-				                       inst.branch_target),
-				           error);
-				return false;
+				ExitBuildFailure(graph, FailureKind::InvalidBranchTarget, UINT32_MAX,
+				                 fmt::format("branch at pc 0x{:08x} targets invalid pc 0x{:08x}",
+				                             inst.pc, inst.branch_target));
 			}
 			labels.insert(inst.branch_target);
 			if (next_pc <= end_pc) {
@@ -1935,21 +1929,19 @@ bool BuildGraph(const Decoder::Program& program, Graph& graph, std::string* erro
 		} else if (inst.opcode == Opcode::S_SETPC_B64) {
 			SetpcTargetInfo target_info;
 			if (!ResolveSetpcTargets(program, i, target_info)) {
-				SetFailure(graph, FailureKind::InvalidBranchTarget, UINT32_MAX,
-				           fmt::format("unsupported dynamic S_SETPC_B64 at pc 0x{:08x}", inst.pc),
-				           error);
-				return false;
+				ExitBuildFailure(
+				    graph, FailureKind::InvalidBranchTarget, UINT32_MAX,
+				    fmt::format("unsupported dynamic S_SETPC_B64 at pc 0x{:08x}", inst.pc));
 			}
 			const auto& target_pcs = target_info.indirect
 			                             ? target_info.target_pcs
 			                             : std::vector<uint32_t> {target_info.target};
 			for (const auto target: target_pcs) {
 				if (!IsValidTarget(target, instruction_pcs, first_pc, end_pc)) {
-					SetFailure(graph, FailureKind::InvalidBranchTarget, UINT32_MAX,
-					           fmt::format("S_SETPC_B64 at pc 0x{:08x} targets invalid pc 0x{:08x}",
-					                       inst.pc, target),
-					           error);
-					return false;
+					ExitBuildFailure(
+					    graph, FailureKind::InvalidBranchTarget, UINT32_MAX,
+					    fmt::format("S_SETPC_B64 at pc 0x{:08x} targets invalid pc 0x{:08x}",
+					                inst.pc, target));
 				}
 				labels.insert(target);
 			}
@@ -1970,10 +1962,9 @@ bool BuildGraph(const Decoder::Program& program, Graph& graph, std::string* erro
 			continue;
 		}
 		if (start != end_pc && !instruction_pcs.contains(start)) {
-			SetFailure(graph, FailureKind::InvalidLabel, UINT32_MAX,
-			           fmt::format("CFG label does not start on an instruction: 0x{:08x}", start),
-			           error);
-			return false;
+			ExitBuildFailure(
+			    graph, FailureKind::InvalidLabel, UINT32_MAX,
+			    fmt::format("CFG label does not start on an instruction: 0x{:08x}", start));
 		}
 
 		BasicBlock block;
@@ -2045,11 +2036,10 @@ bool BuildGraph(const Decoder::Program& program, Graph& graph, std::string* erro
 			block.terminator.true_block = pc_to_block.at(last.branch_target);
 			const auto fallthrough      = pc_to_block.find(next_pc);
 			if (fallthrough == pc_to_block.end()) {
-				SetFailure(graph, FailureKind::MissingFallthrough, block.id,
-				           fmt::format("conditional branch at pc 0x{:08x} has no fallthrough block",
-				                       last.pc),
-				           error);
-				return false;
+				ExitBuildFailure(
+				    graph, FailureKind::MissingFallthrough, block.id,
+				    fmt::format("conditional branch at pc 0x{:08x} has no fallthrough block",
+				                last.pc));
 			}
 			block.terminator.false_block = fallthrough->second;
 		} else {
@@ -2130,27 +2120,26 @@ bool BuildGraph(const Decoder::Program& program, Graph& graph, std::string* erro
 		graph.unsupported = false;
 	}
 
-	return true;
+	return graph;
 }
 
 namespace {
 
-bool StructurizeImpl(Graph& graph, std::string* error) {
+bool StructurizeImpl(Graph& graph) {
 	if (graph.unsupported || graph.irreducible) {
 		if (graph.unsupported_reason.empty()) {
 			graph.unsupported_reason = "unsupported CFG";
 		}
-		SetError(error, graph.unsupported_reason);
 		return false;
 	}
 
-	if (!CanonicalizeNaturalLoops(graph, error)) {
+	if (!CanonicalizeNaturalLoops(graph)) {
 		return false;
 	}
-	if (!SplitSharedMergeBlocks(graph, error)) {
+	if (!SplitSharedMergeBlocks(graph)) {
 		return false;
 	}
-	if (!IsolateSemanticLoopHeaders(graph, error)) {
+	if (!IsolateSemanticLoopHeaders(graph)) {
 		return false;
 	}
 	ClearStructuredTerminators(graph);
@@ -2163,8 +2152,7 @@ bool StructurizeImpl(Graph& graph, std::string* error) {
 			    graph, FailureKind::StructuredControlFlow, header,
 			    fmt::format(
 			        "duplicate structured merge block {} for header {} (already used by header {})",
-			        merge, header, it->second),
-			    error);
+			        merge, header, it->second));
 			return false;
 		}
 		return true;
@@ -2175,16 +2163,14 @@ bool StructurizeImpl(Graph& graph, std::string* error) {
 		if (header == nullptr || loop.merge == UINT32_MAX || loop.continue_block == UINT32_MAX) {
 			SetFailure(
 			    graph, FailureKind::StructuredControlFlow, loop.header,
-			    fmt::format("loop at block {} has no structured merge/continue", loop.header),
-			    error);
+			    fmt::format("loop at block {} has no structured merge/continue", loop.header));
 			return false;
 		}
 		if (header->inst_begin != header->inst_end ||
 		    header->terminator.kind != TerminatorKind::Branch) {
-			SetFailure(graph, FailureKind::StructuredControlFlow, loop.header,
-			           fmt::format("loop header {} is not a dedicated empty control block",
-			                       loop.header),
-			           error);
+			SetFailure(
+			    graph, FailureKind::StructuredControlFlow, loop.header,
+			    fmt::format("loop header {} is not a dedicated empty control block", loop.header));
 			return false;
 		}
 		if (!reserve_merge_block(loop.header, loop.merge)) {
@@ -2207,8 +2193,7 @@ bool StructurizeImpl(Graph& graph, std::string* error) {
 		const auto merge = FindSelectionMerge(graph, block);
 		if (merge == UINT32_MAX) {
 			SetFailure(graph, FailureKind::StructuredControlFlow, block.id,
-			           fmt::format("conditional block {} has no structured merge", block.id),
-			           error);
+			           fmt::format("conditional block {} has no structured merge", block.id));
 			return false;
 		}
 
@@ -2223,16 +2208,15 @@ bool StructurizeImpl(Graph& graph, std::string* error) {
 
 } // namespace
 
-bool Structurize(Graph& graph, std::string* error) {
+bool Structurize(Graph& graph) {
 	Graph original = graph;
-	if (StructurizeImpl(graph, error)) {
+	if (StructurizeImpl(graph)) {
 		return true;
 	}
 
-	Graph       failed_graph = std::move(graph);
-	std::string failed_error = error != nullptr ? *error : std::string {};
-	graph                    = std::move(original);
-	const auto route_budget  = static_cast<uint32_t>(graph.blocks.size());
+	Graph failed_graph      = std::move(graph);
+	graph                   = std::move(original);
+	const auto route_budget = static_cast<uint32_t>(graph.blocks.size());
 	// Apply one route at a time and retry. Eagerly routing every matching diamond can
 	// rewrite unrelated selections that were already structurally valid.
 	for (uint32_t route_variable = 0; route_variable < route_budget; route_variable++) {
@@ -2240,18 +2224,12 @@ bool Structurize(Graph& graph, std::string* error) {
 			break;
 		}
 		Graph routed = graph;
-		if (error != nullptr) {
-			error->clear();
-		}
-		if (StructurizeImpl(routed, error)) {
+		if (StructurizeImpl(routed)) {
 			graph = std::move(routed);
 			return true;
 		}
 	}
 	graph = std::move(failed_graph);
-	if (error != nullptr) {
-		*error = std::move(failed_error);
-	}
 	return false;
 }
 

src/graphics/shader/recompiler/frontend/cfg/ShaderCFG.h

@@ -108,8 +108,8 @@ struct Graph {
 	uint32_t          FindNearestCommonPostDominator(uint32_t block_a, uint32_t block_b) const;
 };
 
-bool        BuildGraph(const Decoder::Program& program, Graph& graph, std::string* error);
-bool        Structurize(Graph& graph, std::string* error);
+Graph       BuildGraph(const Decoder::Program& program);
+bool        Structurize(Graph& graph);
 std::string BranchConditionToString(BranchCondition condition);
 std::string FailureKindToString(FailureKind kind);
 std::string GraphToString(const Graph& graph);

src/graphics/shader/recompiler/frontend/decode/ImageOps.cpp

@@ -357,8 +357,8 @@ void DecodeMimg(uint32_t pc, std::span<const uint32_t> code, uint32_t word_index
 
 	DecodeVectorGpr(vdata, inst.dst);
 	DecodeVectorGpr(vaddr, inst.src0);
-	DecodeScalarSource(srsrc * 4u, pc, inst.src1, nullptr);
-	DecodeScalarSource(ssamp * 4u, pc, inst.src2, nullptr);
+	DecodeScalarSource(srsrc * 4u, pc, inst.src1);
+	DecodeScalarSource(ssamp * 4u, pc, inst.src2);
 	inst.src_count = 3;
 }
 

src/graphics/shader/recompiler/frontend/decode/MemoryOps.cpp

@@ -103,7 +103,11 @@ constexpr MemoryOpcodeInfo DS_OPCODE_LIST[] = {
     {0x4du, Opcode::DS_WRITE_B64, 2, 32},        {0x4eu, Opcode::DS_WRITE2_B64, 4, 32},
     {0x4fu, Opcode::DS_WRITE2ST64_B64, 4, 32},   {0x76u, Opcode::DS_READ_B64, 2, 32},
     {0x77u, Opcode::DS_READ2_B64, 4, 32},        {0x78u, Opcode::DS_READ2ST64_B64, 4, 32},
+    {0xa1u, Opcode::DS_WRITE_B16_D16_HI, 1, 16},
+    {0xa6u, Opcode::DS_READ_U16_D16, 1, 16},
+    {0xa7u, Opcode::DS_READ_U16_D16_HI, 1, 16},
     {0xb0u, Opcode::DS_WRITE_ADDTID_B32, 1, 32}, {0xb1u, Opcode::DS_READ_ADDTID_B32, 1, 32},
+    {0xb3u, Opcode::DS_BPERMUTE_B32, 1, 32},
     {0xdeu, Opcode::DS_WRITE_B96, 3, 32},        {0xdfu, Opcode::DS_WRITE_B128, 4, 32},
     {0xfeu, Opcode::DS_READ_B96, 3, 32},         {0xffu, Opcode::DS_READ_B128, 4, 32},
 };
@@ -146,6 +150,7 @@ bool IsDsWriteOpcode(Opcode opcode) {
 	switch (opcode) {
 		case Opcode::DS_WRITE_B8:
 		case Opcode::DS_WRITE_B16:
+		case Opcode::DS_WRITE_B16_D16_HI:
 		case Opcode::DS_WRITE2_B32:
 		case Opcode::DS_WRITE2ST64_B32:
 		case Opcode::DS_WRITE2_B64:
@@ -191,6 +196,7 @@ uint32_t DsSourceCount(Opcode opcode) {
 		case Opcode::DS_WRITE2ST64_B64:
 		case Opcode::DS_MIN_F32:
 		case Opcode::DS_MAX_F32: return 3u;
+		case Opcode::DS_BPERMUTE_B32: return 2u;
 		case Opcode::DS_READ_ADDTID_B32:
 		case Opcode::DS_CONSUME:
 		case Opcode::DS_APPEND: return 0u;
@@ -235,11 +241,11 @@ void DecodeSmem(uint32_t pc, std::span<const uint32_t> code, uint32_t word_index
 		MarkMemoryUnsupported(inst, Family::SMEM, opcode, "SMEM opcode is not implemented");
 	}
 
-	DecodeScalarDestination(sdst, pc, inst.dst, nullptr);
+	DecodeScalarDestination(sdst, pc, inst.dst);
 	// SMEM encodes SBASE in SGPR pairs. Scalar-buffer loads still use the same
 	// pair index; their descriptor operand consumes four SGPRs from that base.
-	DecodeScalarSource(sbase * 2u, pc, inst.src0, nullptr);
-	DecodeScalarSource(soffset, pc, inst.src1, nullptr);
+	DecodeScalarSource(sbase * 2u, pc, inst.src0);
+	DecodeScalarSource(soffset, pc, inst.src1);
 	inst.src_count = 2;
 }
 
@@ -272,8 +278,8 @@ void DecodeMubuf(uint32_t pc, std::span<const uint32_t> code, uint32_t word_inde
 
 	DecodeVectorGpr(vdata, inst.dst);
 	DecodeVectorGpr(vaddr, inst.src0);
-	DecodeScalarSource(srsrc * 4u, pc, inst.src1, nullptr);
-	DecodeScalarSource(soffset, pc, inst.src2, nullptr);
+	DecodeScalarSource(srsrc * 4u, pc, inst.src1);
+	DecodeScalarSource(soffset, pc, inst.src2);
 	inst.src_count = 3;
 }
 
@@ -310,8 +316,8 @@ void DecodeMtbuf(uint32_t pc, std::span<const uint32_t> code, uint32_t word_inde
 
 	DecodeVectorGpr(vdata, inst.dst);
 	DecodeVectorGpr(vaddr, inst.src0);
-	DecodeScalarSource(srsrc * 4u, pc, inst.src1, nullptr);
-	DecodeScalarSource(soffset, pc, inst.src2, nullptr);
+	DecodeScalarSource(srsrc * 4u, pc, inst.src1);
+	DecodeScalarSource(soffset, pc, inst.src2);
 	inst.src_count = 3;
 }
 
@@ -358,7 +364,7 @@ void DecodeFlat(uint32_t pc, std::span<const uint32_t> code, uint32_t word_index
 		DecodeVectorGpr(addr + 1u, inst.src1);
 		inst.src_count = 2;
 	} else {
-		DecodeScalarSource(saddr, pc, inst.src1, nullptr);
+		DecodeScalarSource(saddr, pc, inst.src1);
 		inst.src_count = 2;
 	}
 }
@@ -391,9 +397,10 @@ void DecodeDs(uint32_t pc, std::span<const uint32_t> code, uint32_t word_index,
 		SetUnsupported(inst, Family::DS, opcode, "DS swizzle FFT mode is not implemented");
 	}
 	if (inst.gds &&
-	    (inst.opcode == Opcode::DS_SWIZZLE_B32 || inst.opcode == Opcode::DS_WRITE_ADDTID_B32 ||
+	    (inst.opcode == Opcode::DS_SWIZZLE_B32 || inst.opcode == Opcode::DS_BPERMUTE_B32 ||
+	     inst.opcode == Opcode::DS_WRITE_ADDTID_B32 ||
 	     inst.opcode == Opcode::DS_READ_ADDTID_B32)) {
-		SetUnsupported(inst, Family::DS, opcode, "DS swizzle/addtid is available only for LDS");
+		SetUnsupported(inst, Family::DS, opcode, "DS lane operation is available only for LDS");
 	}
 	if (inst.opcode == Opcode::DS_WRITE_ADDTID_B32 && data1 != 0u) {
 		SetUnsupported(inst, Family::DS, opcode,
@@ -420,8 +427,17 @@ void DecodeDs(uint32_t pc, std::span<const uint32_t> code, uint32_t word_index,
 	}
 
 	DecodeVectorGpr(vdst, inst.dst);
+	if (inst.opcode == Opcode::DS_READ_U16_D16 ||
+	    inst.opcode == Opcode::DS_READ_U16_D16_HI) {
+		// Native D16 reads update only the selected destination half. Reuse the partial
+		// destination representation so typed IR translation preserves the other half.
+		inst.dst.sdwa_sel = inst.opcode == Opcode::DS_READ_U16_D16 ? 4u : 5u;
+	}
 	DecodeVectorGpr(addr, inst.src0);
 	DecodeVectorGpr(data0, inst.src1);
+	if (inst.opcode == Opcode::DS_WRITE_B16_D16_HI) {
+		inst.src1.sdwa_sel = 5u;
+	}
 	DecodeVectorGpr(data1, inst.src2);
 	inst.src_count = DsSourceCount(inst.opcode);
 }