diff options
Diffstat (limited to '')
27 files changed, 319 insertions, 135 deletions
diff --git a/src/video_core/engines/maxwell_3d.cpp b/src/video_core/engines/maxwell_3d.cpp index 592c28ba3..95ba4f76c 100644 --- a/src/video_core/engines/maxwell_3d.cpp +++ b/src/video_core/engines/maxwell_3d.cpp @@ -586,14 +586,22 @@ void Maxwell3D::ProcessQueryCondition() { } void Maxwell3D::ProcessCounterReset() { - switch (regs.clear_report_value) { - case Regs::ClearReport::ZPassPixelCount: - rasterizer->ResetCounter(VideoCommon::QueryType::ZPassPixelCount64); - break; - default: - LOG_DEBUG(Render_OpenGL, "Unimplemented counter reset={}", regs.clear_report_value); - break; - } + const auto query_type = [clear_report = regs.clear_report_value]() { + switch (clear_report) { + case Tegra::Engines::Maxwell3D::Regs::ClearReport::ZPassPixelCount: + return VideoCommon::QueryType::ZPassPixelCount64; + case Tegra::Engines::Maxwell3D::Regs::ClearReport::StreamingPrimitivesSucceeded: + return VideoCommon::QueryType::StreamingPrimitivesSucceeded; + case Tegra::Engines::Maxwell3D::Regs::ClearReport::PrimitivesGenerated: + return VideoCommon::QueryType::PrimitivesGenerated; + case Tegra::Engines::Maxwell3D::Regs::ClearReport::VtgPrimitivesOut: + return VideoCommon::QueryType::VtgPrimitivesOut; + default: + LOG_DEBUG(HW_GPU, "Unimplemented counter reset={}", clear_report); + return VideoCommon::QueryType::Payload; + } + }(); + rasterizer->ResetCounter(query_type); } void Maxwell3D::ProcessSyncPoint() { diff --git a/src/video_core/engines/maxwell_dma.cpp b/src/video_core/engines/maxwell_dma.cpp index 422d4d859..56fbff306 100644 --- a/src/video_core/engines/maxwell_dma.cpp +++ b/src/video_core/engines/maxwell_dma.cpp @@ -228,7 +228,7 @@ void MaxwellDMA::CopyBlockLinearToPitch() { Core::Memory::GpuGuestMemory<u8, Core::Memory::GuestMemoryFlags::SafeRead> tmp_read_buffer( memory_manager, src_operand.address, src_size, &read_buffer); - Core::Memory::GpuGuestMemoryScoped<u8, Core::Memory::GuestMemoryFlags::SafeReadCachedWrite> + Core::Memory::GpuGuestMemoryScoped<u8, Core::Memory::GuestMemoryFlags::UnsafeReadCachedWrite> tmp_write_buffer(memory_manager, dst_operand.address, dst_size, &write_buffer); UnswizzleSubrect(tmp_write_buffer, tmp_read_buffer, bytes_per_pixel, width, height, depth, @@ -292,7 +292,7 @@ void MaxwellDMA::CopyPitchToBlockLinear() { GPUVAddr dst_addr = regs.offset_out; Core::Memory::GpuGuestMemory<u8, Core::Memory::GuestMemoryFlags::SafeRead> tmp_read_buffer( memory_manager, src_addr, src_size, &read_buffer); - Core::Memory::GpuGuestMemoryScoped<u8, Core::Memory::GuestMemoryFlags::SafeReadCachedWrite> + Core::Memory::GpuGuestMemoryScoped<u8, Core::Memory::GuestMemoryFlags::UnsafeReadCachedWrite> tmp_write_buffer(memory_manager, dst_addr, dst_size, &write_buffer); // If the input is linear and the output is tiled, swizzle the input and copy it over. diff --git a/src/video_core/macro/macro_hle.cpp b/src/video_core/macro/macro_hle.cpp index 046c8085e..46e853e04 100644 --- a/src/video_core/macro/macro_hle.cpp +++ b/src/video_core/macro/macro_hle.cpp @@ -327,12 +327,13 @@ public: explicit HLE_DrawIndirectByteCount(Maxwell3D& maxwell3d_) : HLEMacroImpl(maxwell3d_) {} void Execute(const std::vector<u32>& parameters, [[maybe_unused]] u32 method) override { + const bool force = maxwell3d.Rasterizer().HasDrawTransformFeedback(); + auto topology = static_cast<Maxwell3D::Regs::PrimitiveTopology>(parameters[0] & 0xFFFFU); - if (!maxwell3d.AnyParametersDirty() || !IsTopologySafe(topology)) { + if (!force && (!maxwell3d.AnyParametersDirty() || !IsTopologySafe(topology))) { Fallback(parameters); return; } - auto& params = maxwell3d.draw_manager->GetIndirectParams(); params.is_byte_count = true; params.is_indexed = false; @@ -503,6 +504,8 @@ public: maxwell3d.CallMethod(static_cast<size_t>(MAXWELL3D_REG_INDEX(launch_dma)), 0x1011, true); maxwell3d.CallMethod(static_cast<size_t>(MAXWELL3D_REG_INDEX(inline_data)), regs.transform_feedback.controls[0].stride, true); + + maxwell3d.Rasterizer().RegisterTransformFeedback(regs.upload.dest.Address()); } }; diff --git a/src/video_core/query_cache.h b/src/video_core/query_cache.h index 9fcaeeac7..a64404ce4 100644 --- a/src/video_core/query_cache.h +++ b/src/video_core/query_cache.h @@ -28,8 +28,11 @@ namespace VideoCore { enum class QueryType { SamplesPassed, + PrimitivesGenerated, + TfbPrimitivesWritten, + Count, }; -constexpr std::size_t NumQueryTypes = 1; +constexpr std::size_t NumQueryTypes = static_cast<size_t>(QueryType::Count); } // namespace VideoCore namespace VideoCommon { @@ -44,15 +47,6 @@ public: explicit CounterStreamBase(QueryCache& cache_, VideoCore::QueryType type_) : cache{cache_}, type{type_} {} - /// Updates the state of the stream, enabling or disabling as needed. - void Update(bool enabled) { - if (enabled) { - Enable(); - } else { - Disable(); - } - } - /// Resets the stream to zero. It doesn't disable the query after resetting. void Reset() { if (current) { @@ -80,7 +74,6 @@ public: return current != nullptr; } -private: /// Enables the stream. void Enable() { if (current) { @@ -97,6 +90,7 @@ private: last = std::exchange(current, nullptr); } +private: QueryCache& cache; const VideoCore::QueryType type; @@ -112,8 +106,14 @@ public: : rasterizer{rasterizer_}, // Use reinterpret_cast instead of static_cast as workaround for // UBSan bug (https://github.com/llvm/llvm-project/issues/59060) - cpu_memory{cpu_memory_}, streams{{CounterStream{reinterpret_cast<QueryCache&>(*this), - VideoCore::QueryType::SamplesPassed}}} { + cpu_memory{cpu_memory_}, streams{{ + {CounterStream{reinterpret_cast<QueryCache&>(*this), + VideoCore::QueryType::SamplesPassed}}, + {CounterStream{reinterpret_cast<QueryCache&>(*this), + VideoCore::QueryType::PrimitivesGenerated}}, + {CounterStream{reinterpret_cast<QueryCache&>(*this), + VideoCore::QueryType::TfbPrimitivesWritten}}, + }} { (void)slot_async_jobs.insert(); // Null value } @@ -157,12 +157,11 @@ public: AsyncFlushQuery(query, timestamp, lock); } - /// Updates counters from GPU state. Expected to be called once per draw, clear or dispatch. - void UpdateCounters() { + /// Enables all available GPU counters + void EnableCounters() { std::unique_lock lock{mutex}; - if (maxwell3d) { - const auto& regs = maxwell3d->regs; - Stream(VideoCore::QueryType::SamplesPassed).Update(regs.zpass_pixel_count_enable); + for (auto& stream : streams) { + stream.Enable(); } } @@ -176,7 +175,7 @@ public: void DisableStreams() { std::unique_lock lock{mutex}; for (auto& stream : streams) { - stream.Update(false); + stream.Disable(); } } @@ -353,7 +352,7 @@ private: std::shared_ptr<std::vector<AsyncJobId>> uncommitted_flushes{}; std::list<std::shared_ptr<std::vector<AsyncJobId>>> committed_flushes; -}; +}; // namespace VideoCommon template <class QueryCache, class HostCounter> class HostCounterBase { diff --git a/src/video_core/rasterizer_interface.h b/src/video_core/rasterizer_interface.h index af1469147..49224ca85 100644 --- a/src/video_core/rasterizer_interface.h +++ b/src/video_core/rasterizer_interface.h @@ -173,5 +173,13 @@ public: virtual void BindChannel(Tegra::Control::ChannelState& channel) {} virtual void ReleaseChannel(s32 channel_id) {} + + /// Register the address as a Transform Feedback Object + virtual void RegisterTransformFeedback(GPUVAddr tfb_object_addr) {} + + /// Returns true when the rasterizer has Draw Transform Feedback capabilities + virtual bool HasDrawTransformFeedback() { + return false; + } }; } // namespace VideoCore diff --git a/src/video_core/renderer_opengl/gl_buffer_cache.cpp b/src/video_core/renderer_opengl/gl_buffer_cache.cpp index b787b6994..517ac14dd 100644 --- a/src/video_core/renderer_opengl/gl_buffer_cache.cpp +++ b/src/video_core/renderer_opengl/gl_buffer_cache.cpp @@ -376,4 +376,15 @@ void BufferCacheRuntime::BindImageBuffer(Buffer& buffer, u32 offset, u32 size, P *image_handles++ = buffer.View(offset, size, format); } +void BufferCacheRuntime::BindTransformFeedbackObject(GPUVAddr tfb_object_addr) { + OGLTransformFeedback& tfb_object = tfb_objects[tfb_object_addr]; + tfb_object.Create(); + glBindTransformFeedback(GL_TRANSFORM_FEEDBACK, tfb_object.handle); +} + +GLuint BufferCacheRuntime::GetTransformFeedbackObject(GPUVAddr tfb_object_addr) { + ASSERT(tfb_objects.contains(tfb_object_addr)); + return tfb_objects[tfb_object_addr].handle; +} + } // namespace OpenGL diff --git a/src/video_core/renderer_opengl/gl_buffer_cache.h b/src/video_core/renderer_opengl/gl_buffer_cache.h index 1e8708f59..2c18de166 100644 --- a/src/video_core/renderer_opengl/gl_buffer_cache.h +++ b/src/video_core/renderer_opengl/gl_buffer_cache.h @@ -5,6 +5,7 @@ #include <array> #include <span> +#include <unordered_map> #include "common/common_types.h" #include "video_core/buffer_cache/buffer_cache_base.h" @@ -121,6 +122,9 @@ public: void BindImageBuffer(Buffer& buffer, u32 offset, u32 size, VideoCore::Surface::PixelFormat format); + void BindTransformFeedbackObject(GPUVAddr tfb_object_addr); + GLuint GetTransformFeedbackObject(GPUVAddr tfb_object_addr); + u64 GetDeviceMemoryUsage() const; void BindFastUniformBuffer(size_t stage, u32 binding_index, u32 size) { @@ -233,6 +237,7 @@ private: u32 index_buffer_offset = 0; u64 device_access_memory; + std::unordered_map<GPUVAddr, OGLTransformFeedback> tfb_objects; }; struct BufferCacheParams { diff --git a/src/video_core/renderer_opengl/gl_query_cache.cpp b/src/video_core/renderer_opengl/gl_query_cache.cpp index ec142d48e..fef7360ed 100644 --- a/src/video_core/renderer_opengl/gl_query_cache.cpp +++ b/src/video_core/renderer_opengl/gl_query_cache.cpp @@ -18,16 +18,27 @@ namespace OpenGL { namespace { -constexpr std::array<GLenum, VideoCore::NumQueryTypes> QueryTargets = {GL_SAMPLES_PASSED}; - constexpr GLenum GetTarget(VideoCore::QueryType type) { - return QueryTargets[static_cast<std::size_t>(type)]; + switch (type) { + case VideoCore::QueryType::SamplesPassed: + return GL_SAMPLES_PASSED; + case VideoCore::QueryType::PrimitivesGenerated: + return GL_PRIMITIVES_GENERATED; + case VideoCore::QueryType::TfbPrimitivesWritten: + return GL_TRANSFORM_FEEDBACK_PRIMITIVES_WRITTEN; + default: + break; + } + UNIMPLEMENTED_MSG("Query type {}", type); + return 0; } } // Anonymous namespace QueryCache::QueryCache(RasterizerOpenGL& rasterizer_, Core::Memory::Memory& cpu_memory_) - : QueryCacheLegacy(rasterizer_, cpu_memory_), gl_rasterizer{rasterizer_} {} + : QueryCacheLegacy(rasterizer_, cpu_memory_), gl_rasterizer{rasterizer_} { + EnableCounters(); +} QueryCache::~QueryCache() = default; @@ -103,13 +114,13 @@ u64 CachedQuery::Flush([[maybe_unused]] bool async) { auto& stream = cache->Stream(type); const bool slice_counter = WaitPending() && stream.IsEnabled(); if (slice_counter) { - stream.Update(false); + stream.Disable(); } auto result = VideoCommon::CachedQueryBase<HostCounter>::Flush(); if (slice_counter) { - stream.Update(true); + stream.Enable(); } return result; diff --git a/src/video_core/renderer_opengl/gl_rasterizer.cpp b/src/video_core/renderer_opengl/gl_rasterizer.cpp index 4832c03c5..7a5fad735 100644 --- a/src/video_core/renderer_opengl/gl_rasterizer.cpp +++ b/src/video_core/renderer_opengl/gl_rasterizer.cpp @@ -51,6 +51,22 @@ constexpr size_t NUM_SUPPORTED_VERTEX_ATTRIBUTES = 16; void oglEnable(GLenum cap, bool state) { (state ? glEnable : glDisable)(cap); } + +std::optional<VideoCore::QueryType> MaxwellToVideoCoreQuery(VideoCommon::QueryType type) { + switch (type) { + case VideoCommon::QueryType::PrimitivesGenerated: + case VideoCommon::QueryType::VtgPrimitivesOut: + return VideoCore::QueryType::PrimitivesGenerated; + case VideoCommon::QueryType::ZPassPixelCount64: + return VideoCore::QueryType::SamplesPassed; + case VideoCommon::QueryType::StreamingPrimitivesSucceeded: + // case VideoCommon::QueryType::StreamingByteCount: + // TODO: StreamingByteCount = StreamingPrimitivesSucceeded * num_verts * vert_stride + return VideoCore::QueryType::TfbPrimitivesWritten; + default: + return std::nullopt; + } +} } // Anonymous namespace RasterizerOpenGL::RasterizerOpenGL(Core::Frontend::EmuWindow& emu_window_, Tegra::GPU& gpu_, @@ -216,7 +232,6 @@ void RasterizerOpenGL::PrepareDraw(bool is_indexed, Func&& draw_func) { SCOPE_EXIT({ gpu.TickWork(); }); gpu_memory->FlushCaching(); - query_cache.UpdateCounters(); GraphicsPipeline* const pipeline{shader_cache.CurrentGraphicsPipeline()}; if (!pipeline) { @@ -294,6 +309,13 @@ void RasterizerOpenGL::DrawIndirect() { const auto& params = maxwell3d->draw_manager->GetIndirectParams(); buffer_cache.SetDrawIndirect(¶ms); PrepareDraw(params.is_indexed, [this, ¶ms](GLenum primitive_mode) { + if (params.is_byte_count) { + const GPUVAddr tfb_object_base_addr = params.indirect_start_address - 4U; + const GLuint tfb_object = + buffer_cache_runtime.GetTransformFeedbackObject(tfb_object_base_addr); + glDrawTransformFeedback(primitive_mode, tfb_object); + return; + } const auto [buffer, offset] = buffer_cache.GetDrawIndirectBuffer(); const GLvoid* const gl_offset = reinterpret_cast<const GLvoid*>(static_cast<uintptr_t>(offset)); @@ -334,7 +356,6 @@ void RasterizerOpenGL::DrawTexture() { MICROPROFILE_SCOPE(OpenGL_Drawing); SCOPE_EXIT({ gpu.TickWork(); }); - query_cache.UpdateCounters(); texture_cache.SynchronizeGraphicsDescriptors(); texture_cache.UpdateRenderTargets(false); @@ -401,21 +422,28 @@ void RasterizerOpenGL::DispatchCompute() { } void RasterizerOpenGL::ResetCounter(VideoCommon::QueryType type) { - if (type == VideoCommon::QueryType::ZPassPixelCount64) { - query_cache.ResetCounter(VideoCore::QueryType::SamplesPassed); + const auto query_cache_type = MaxwellToVideoCoreQuery(type); + if (!query_cache_type.has_value()) { + UNIMPLEMENTED_IF_MSG(type != VideoCommon::QueryType::Payload, "Reset query type: {}", type); + return; } + query_cache.ResetCounter(*query_cache_type); } void RasterizerOpenGL::Query(GPUVAddr gpu_addr, VideoCommon::QueryType type, VideoCommon::QueryPropertiesFlags flags, u32 payload, u32 subreport) { - if (type == VideoCommon::QueryType::ZPassPixelCount64) { - if (True(flags & VideoCommon::QueryPropertiesFlags::HasTimeout)) { - query_cache.Query(gpu_addr, VideoCore::QueryType::SamplesPassed, {gpu.GetTicks()}); - } else { - query_cache.Query(gpu_addr, VideoCore::QueryType::SamplesPassed, std::nullopt); - } - return; + const auto query_cache_type = MaxwellToVideoCoreQuery(type); + if (!query_cache_type.has_value()) { + return QueryFallback(gpu_addr, type, flags, payload, subreport); } + const bool has_timeout = True(flags & VideoCommon::QueryPropertiesFlags::HasTimeout); + const auto timestamp = has_timeout ? std::optional<u64>{gpu.GetTicks()} : std::nullopt; + query_cache.Query(gpu_addr, *query_cache_type, timestamp); +} + +void RasterizerOpenGL::QueryFallback(GPUVAddr gpu_addr, VideoCommon::QueryType type, + VideoCommon::QueryPropertiesFlags flags, u32 payload, + u32 subreport) { if (type != VideoCommon::QueryType::Payload) { payload = 1u; } @@ -1350,6 +1378,10 @@ void RasterizerOpenGL::ReleaseChannel(s32 channel_id) { query_cache.EraseChannel(channel_id); } +void RasterizerOpenGL::RegisterTransformFeedback(GPUVAddr tfb_object_addr) { + buffer_cache_runtime.BindTransformFeedbackObject(tfb_object_addr); +} + AccelerateDMA::AccelerateDMA(BufferCache& buffer_cache_, TextureCache& texture_cache_) : buffer_cache{buffer_cache_}, texture_cache{texture_cache_} {} diff --git a/src/video_core/renderer_opengl/gl_rasterizer.h b/src/video_core/renderer_opengl/gl_rasterizer.h index ceffe1f1e..ce3460938 100644 --- a/src/video_core/renderer_opengl/gl_rasterizer.h +++ b/src/video_core/renderer_opengl/gl_rasterizer.h @@ -139,6 +139,12 @@ public: void ReleaseChannel(s32 channel_id) override; + void RegisterTransformFeedback(GPUVAddr tfb_object_addr) override; + + bool HasDrawTransformFeedback() override { + return true; + } + private: static constexpr size_t MAX_TEXTURES = 192; static constexpr size_t MAX_IMAGES = 48; @@ -225,6 +231,9 @@ private: /// End a transform feedback void EndTransformFeedback(); + void QueryFallback(GPUVAddr gpu_addr, VideoCommon::QueryType type, + VideoCommon::QueryPropertiesFlags flags, u32 payload, u32 subreport); + Tegra::GPU& gpu; const Device& device; diff --git a/src/video_core/renderer_opengl/gl_resource_manager.cpp b/src/video_core/renderer_opengl/gl_resource_manager.cpp index eae8fd110..1d2c9b70a 100644 --- a/src/video_core/renderer_opengl/gl_resource_manager.cpp +++ b/src/video_core/renderer_opengl/gl_resource_manager.cpp @@ -207,4 +207,21 @@ void OGLQuery::Release() { handle = 0; } +void OGLTransformFeedback::Create() { + if (handle != 0) + return; + + MICROPROFILE_SCOPE(OpenGL_ResourceCreation); + glCreateTransformFeedbacks(1, &handle); +} + +void OGLTransformFeedback::Release() { + if (handle == 0) + return; + + MICROPROFILE_SCOPE(OpenGL_ResourceDeletion); + glDeleteTransformFeedbacks(1, &handle); + handle = 0; +} + } // namespace OpenGL diff --git a/src/video_core/renderer_opengl/gl_resource_manager.h b/src/video_core/renderer_opengl/gl_resource_manager.h index 77362acd2..6ca8227bd 100644 --- a/src/video_core/renderer_opengl/gl_resource_manager.h +++ b/src/video_core/renderer_opengl/gl_resource_manager.h @@ -323,4 +323,31 @@ public: GLuint handle = 0; }; +class OGLTransformFeedback final { +public: + YUZU_NON_COPYABLE(OGLTransformFeedback); + + OGLTransformFeedback() = default; + + OGLTransformFeedback(OGLTransformFeedback&& o) noexcept : handle(std::exchange(o.handle, 0)) {} + + ~OGLTransformFeedback() { + Release(); + } + + OGLTransformFeedback& operator=(OGLTransformFeedback&& o) noexcept { + Release(); + handle = std::exchange(o.handle, 0); + return *this; + } + + /// Creates a new internal OpenGL resource and stores the handle + void Create(); + + /// Deletes the internal OpenGL resource + void Release(); + + GLuint handle = 0; +}; + } // namespace OpenGL diff --git a/src/video_core/renderer_opengl/gl_shader_cache.cpp b/src/video_core/renderer_opengl/gl_shader_cache.cpp index 26f2d0ea7..30df41b7d 100644 --- a/src/video_core/renderer_opengl/gl_shader_cache.cpp +++ b/src/video_core/renderer_opengl/gl_shader_cache.cpp @@ -51,7 +51,7 @@ using VideoCommon::LoadPipelines; using VideoCommon::SerializePipeline; using Context = ShaderContext::Context; -constexpr u32 CACHE_VERSION = 9; +constexpr u32 CACHE_VERSION = 10; template <typename Container> auto MakeSpan(Container& container) { @@ -233,6 +233,7 @@ ShaderCache::ShaderCache(RasterizerOpenGL& rasterizer_, Core::Frontend::EmuWindo .ignore_nan_fp_comparisons = true, .gl_max_compute_smem_size = device.GetMaxComputeSharedMemorySize(), .min_ssbo_alignment = device.GetShaderStorageBufferAlignment(), + .max_user_clip_distances = 8, }, host_info{ .support_float64 = true, diff --git a/src/video_core/renderer_vulkan/vk_buffer_cache.cpp b/src/video_core/renderer_vulkan/vk_buffer_cache.cpp index 5958f52f7..3c61799fa 100644 --- a/src/video_core/renderer_vulkan/vk_buffer_cache.cpp +++ b/src/video_core/renderer_vulkan/vk_buffer_cache.cpp @@ -78,8 +78,15 @@ vk::Buffer CreateBuffer(const Device& device, const MemoryAllocator& memory_allo } } // Anonymous namespace -Buffer::Buffer(BufferCacheRuntime&, VideoCommon::NullBufferParams null_params) - : VideoCommon::BufferBase<VideoCore::RasterizerInterface>(null_params), tracker{4096} {} +Buffer::Buffer(BufferCacheRuntime& runtime, VideoCommon::NullBufferParams null_params) + : VideoCommon::BufferBase<VideoCore::RasterizerInterface>(null_params), tracker{4096} { + if (runtime.device.HasNullDescriptor()) { + return; + } + device = &runtime.device; + buffer = runtime.CreateNullBuffer(); + is_null = true; +} Buffer::Buffer(BufferCacheRuntime& runtime, VideoCore::RasterizerInterface& rasterizer_, VAddr cpu_addr_, u64 size_bytes_) @@ -93,8 +100,12 @@ Buffer::Buffer(BufferCacheRuntime& runtime, VideoCore::RasterizerInterface& rast VkBufferView Buffer::View(u32 offset, u32 size, VideoCore::Surface::PixelFormat format) { if (!device) { - // Null buffer, return a null descriptor + // Null buffer supported, return a null descriptor return VK_NULL_HANDLE; + } else if (is_null) { + // Null buffer not supported, adjust offset and size + offset = 0; + size = 0; } const auto it{std::ranges::find_if(views, [offset, size, format](const BufferView& view) { return offset == view.offset && size == view.size && format == view.format; @@ -563,22 +574,27 @@ void BufferCacheRuntime::BindVertexBuffers(VideoCommon::HostBindings<Buffer>& bi } buffer_handles.push_back(handle); } + const u32 device_max = device.GetMaxVertexInputBindings(); + const u32 min_binding = std::min(bindings.min_index, device_max); + const u32 max_binding = std::min(bindings.max_index, device_max); + const u32 binding_count = max_binding - min_binding; + if (binding_count == 0) { + return; + } if (device.IsExtExtendedDynamicStateSupported()) { - scheduler.Record([this, bindings_ = std::move(bindings), - buffer_handles_ = std::move(buffer_handles)](vk::CommandBuffer cmdbuf) { - cmdbuf.BindVertexBuffers2EXT(bindings_.min_index, - std::min(bindings_.max_index - bindings_.min_index, - device.GetMaxVertexInputBindings()), - buffer_handles_.data(), bindings_.offsets.data(), - bindings_.sizes.data(), bindings_.strides.data()); + scheduler.Record([bindings_ = std::move(bindings), + buffer_handles_ = std::move(buffer_handles), + binding_count](vk::CommandBuffer cmdbuf) { + cmdbuf.BindVertexBuffers2EXT(bindings_.min_index, binding_count, buffer_handles_.data(), + bindings_.offsets.data(), bindings_.sizes.data(), + bindings_.strides.data()); }); } else { - scheduler.Record([this, bindings_ = std::move(bindings), - buffer_handles_ = std::move(buffer_handles)](vk::CommandBuffer cmdbuf) { - cmdbuf.BindVertexBuffers(bindings_.min_index, - std::min(bindings_.max_index - bindings_.min_index, - device.GetMaxVertexInputBindings()), - buffer_handles_.data(), bindings_.offsets.data()); + scheduler.Record([bindings_ = std::move(bindings), + buffer_handles_ = std::move(buffer_handles), + binding_count](vk::CommandBuffer cmdbuf) { + cmdbuf.BindVertexBuffers(bindings_.min_index, binding_count, buffer_handles_.data(), + bindings_.offsets.data()); }); } } @@ -622,9 +638,12 @@ void BufferCacheRuntime::BindTransformFeedbackBuffers(VideoCommon::HostBindings< } void BufferCacheRuntime::ReserveNullBuffer() { - if (null_buffer) { - return; + if (!null_buffer) { + null_buffer = CreateNullBuffer(); } +} + +vk::Buffer BufferCacheRuntime::CreateNullBuffer() { VkBufferCreateInfo create_info{ .sType = VK_STRUCTURE_TYPE_BUFFER_CREATE_INFO, .pNext = nullptr, @@ -639,15 +658,17 @@ void BufferCacheRuntime::ReserveNullBuffer() { if (device.IsExtTransformFeedbackSupported()) { create_info.usage |= VK_BUFFER_USAGE_TRANSFORM_FEEDBACK_BUFFER_BIT_EXT; } - null_buffer = memory_allocator.CreateBuffer(create_info, MemoryUsage::DeviceLocal); + vk::Buffer ret = memory_allocator.CreateBuffer(create_info, MemoryUsage::DeviceLocal); if (device.HasDebuggingToolAttached()) { - null_buffer.SetObjectNameEXT("Null buffer"); + ret.SetObjectNameEXT("Null buffer"); } scheduler.RequestOutsideRenderPassOperationContext(); - scheduler.Record([buffer = *null_buffer](vk::CommandBuffer cmdbuf) { + scheduler.Record([buffer = *ret](vk::CommandBuffer cmdbuf) { cmdbuf.FillBuffer(buffer, 0, VK_WHOLE_SIZE, 0); }); + + return ret; } } // namespace Vulkan diff --git a/src/video_core/renderer_vulkan/vk_buffer_cache.h b/src/video_core/renderer_vulkan/vk_buffer_cache.h index 0b3fbd6d0..dc300d7cb 100644 --- a/src/video_core/renderer_vulkan/vk_buffer_cache.h +++ b/src/video_core/renderer_vulkan/vk_buffer_cache.h @@ -63,6 +63,7 @@ private: vk::Buffer buffer; std::vector<BufferView> views; VideoCommon::UsageTracker tracker; + bool is_null{}; }; class QuadArrayIndexBuffer; @@ -151,6 +152,7 @@ private: } void ReserveNullBuffer(); + vk::Buffer CreateNullBuffer(); const Device& device; MemoryAllocator& memory_allocator; diff --git a/src/video_core/renderer_vulkan/vk_pipeline_cache.cpp b/src/video_core/renderer_vulkan/vk_pipeline_cache.cpp index 2a13b2a72..d1841198d 100644 --- a/src/video_core/renderer_vulkan/vk_pipeline_cache.cpp +++ b/src/video_core/renderer_vulkan/vk_pipeline_cache.cpp @@ -54,7 +54,7 @@ using VideoCommon::FileEnvironment; using VideoCommon::GenericEnvironment; using VideoCommon::GraphicsEnvironment; -constexpr u32 CACHE_VERSION = 10; +constexpr u32 CACHE_VERSION = 11; constexpr std::array<char, 8> VULKAN_CACHE_MAGIC_NUMBER{'y', 'u', 'z', 'u', 'v', 'k', 'c', 'h'}; template <typename Container> @@ -374,6 +374,7 @@ PipelineCache::PipelineCache(RasterizerVulkan& rasterizer_, const Device& device .has_broken_robust = device.IsNvidia() && device.GetNvidiaArch() <= NvidiaArchitecture::Arch_Pascal, .min_ssbo_alignment = device.GetStorageBufferAlignment(), + .max_user_clip_distances = device.GetMaxUserClipDistances(), }; host_info = Shader::HostTranslateInfo{ diff --git a/src/video_core/renderer_vulkan/vk_present_manager.cpp b/src/video_core/renderer_vulkan/vk_present_manager.cpp index 5e7518d96..792ed9615 100644 --- a/src/video_core/renderer_vulkan/vk_present_manager.cpp +++ b/src/video_core/renderer_vulkan/vk_present_manager.cpp @@ -329,7 +329,7 @@ void PresentManager::CopyToSwapchainImpl(Frame* frame) { // to account for that. const bool is_suboptimal = swapchain.NeedsRecreation(); const bool size_changed = - swapchain.GetWidth() != frame->width || swapchain.GetHeight() != frame->height; + swapchain.GetWidth() < frame->width || swapchain.GetHeight() < frame->height; if (is_suboptimal || size_changed) { RecreateSwapchain(frame); } diff --git a/src/video_core/renderer_vulkan/vk_query_cache.cpp b/src/video_core/renderer_vulkan/vk_query_cache.cpp index 078777cdd..95954ade7 100644 --- a/src/video_core/renderer_vulkan/vk_query_cache.cpp +++ b/src/video_core/renderer_vulkan/vk_query_cache.cpp @@ -289,12 +289,15 @@ public: } if (has_multi_queries) { - size_t intermediary_buffer_index = ObtainBuffer<false>(num_slots_used); + const size_t min_accumulation_limit = + std::min(first_accumulation_checkpoint, num_slots_used); + const size_t max_accumulation_limit = + std::max(last_accumulation_checkpoint, num_slots_used); + const size_t intermediary_buffer_index = ObtainBuffer<false>(num_slots_used); resolve_buffers.push_back(intermediary_buffer_index); queries_prefix_scan_pass->Run(*accumulation_buffer, *buffers[intermediary_buffer_index], *buffers[resolve_buffer_index], num_slots_used, - std::min(first_accumulation_checkpoint, num_slots_used), - last_accumulation_checkpoint); + min_accumulation_limit, max_accumulation_limit); } else { scheduler.RequestOutsideRenderPassOperationContext(); diff --git a/src/video_core/renderer_vulkan/vk_rasterizer.cpp b/src/video_core/renderer_vulkan/vk_rasterizer.cpp index 59829c88b..241fc34be 100644 --- a/src/video_core/renderer_vulkan/vk_rasterizer.cpp +++ b/src/video_core/renderer_vulkan/vk_rasterizer.cpp @@ -485,6 +485,10 @@ void RasterizerVulkan::DispatchCompute() { } void RasterizerVulkan::ResetCounter(VideoCommon::QueryType type) { + if (type != VideoCommon::QueryType::ZPassPixelCount64) { + LOG_DEBUG(Render_Vulkan, "Unimplemented counter reset={}", type); + return; + } query_cache.CounterReset(type); } diff --git a/src/video_core/shader_environment.cpp b/src/video_core/shader_environment.cpp index 4edbe5700..492440ac4 100644 --- a/src/video_core/shader_environment.cpp +++ b/src/video_core/shader_environment.cpp @@ -62,23 +62,9 @@ static Shader::TextureType ConvertTextureType(const Tegra::Texture::TICEntry& en } static Shader::TexturePixelFormat ConvertTexturePixelFormat(const Tegra::Texture::TICEntry& entry) { - switch (PixelFormatFromTextureInfo(entry.format, entry.r_type, entry.g_type, entry.b_type, - entry.a_type, entry.srgb_conversion)) { - case VideoCore::Surface::PixelFormat::A8B8G8R8_SNORM: - return Shader::TexturePixelFormat::A8B8G8R8_SNORM; - case VideoCore::Surface::PixelFormat::R8_SNORM: - return Shader::TexturePixelFormat::R8_SNORM; - case VideoCore::Surface::PixelFormat::R8G8_SNORM: - return Shader::TexturePixelFormat::R8G8_SNORM; - case VideoCore::Surface::PixelFormat::R16G16B16A16_SNORM: - return Shader::TexturePixelFormat::R16G16B16A16_SNORM; - case VideoCore::Surface::PixelFormat::R16G16_SNORM: - return Shader::TexturePixelFormat::R16G16_SNORM; - case VideoCore::Surface::PixelFormat::R16_SNORM: - return Shader::TexturePixelFormat::R16_SNORM; - default: - return Shader::TexturePixelFormat::OTHER; - } + return static_cast<Shader::TexturePixelFormat>( + PixelFormatFromTextureInfo(entry.format, entry.r_type, entry.g_type, entry.b_type, + entry.a_type, entry.srgb_conversion)); } static std::string_view StageToPrefix(Shader::Stage stage) { @@ -398,6 +384,11 @@ Shader::TexturePixelFormat GraphicsEnvironment::ReadTexturePixelFormat(u32 handl return result; } +bool GraphicsEnvironment::IsTexturePixelFormatInteger(u32 handle) { + return VideoCore::Surface::IsPixelFormatInteger( + static_cast<VideoCore::Surface::PixelFormat>(ReadTexturePixelFormat(handle))); +} + u32 GraphicsEnvironment::ReadViewportTransformState() { const auto& regs{maxwell3d->regs}; viewport_transform_state = regs.viewport_scale_offset_enabled; @@ -448,6 +439,11 @@ Shader::TexturePixelFormat ComputeEnvironment::ReadTexturePixelFormat(u32 handle return result; } +bool ComputeEnvironment::IsTexturePixelFormatInteger(u32 handle) { + return VideoCore::Surface::IsPixelFormatInteger( + static_cast<VideoCore::Surface::PixelFormat>(ReadTexturePixelFormat(handle))); +} + u32 ComputeEnvironment::ReadViewportTransformState() { return viewport_transform_state; } @@ -551,6 +547,11 @@ Shader::TexturePixelFormat FileEnvironment::ReadTexturePixelFormat(u32 handle) { return it->second; } +bool FileEnvironment::IsTexturePixelFormatInteger(u32 handle) { + return VideoCore::Surface::IsPixelFormatInteger( + static_cast<VideoCore::Surface::PixelFormat>(ReadTexturePixelFormat(handle))); +} + u32 FileEnvironment::ReadViewportTransformState() { return viewport_transform_state; } diff --git a/src/video_core/shader_environment.h b/src/video_core/shader_environment.h index b90f3d44e..6b372e336 100644 --- a/src/video_core/shader_environment.h +++ b/src/video_core/shader_environment.h @@ -115,6 +115,8 @@ public: Shader::TexturePixelFormat ReadTexturePixelFormat(u32 handle) override; + bool IsTexturePixelFormatInteger(u32 handle) override; + u32 ReadViewportTransformState() override; std::optional<Shader::ReplaceConstant> GetReplaceConstBuffer(u32 bank, u32 offset) override; @@ -139,6 +141,8 @@ public: Shader::TexturePixelFormat ReadTexturePixelFormat(u32 handle) override; + bool IsTexturePixelFormatInteger(u32 handle) override; + u32 ReadViewportTransformState() override; std::optional<Shader::ReplaceConstant> GetReplaceConstBuffer( @@ -171,6 +175,8 @@ public: [[nodiscard]] Shader::TexturePixelFormat ReadTexturePixelFormat(u32 handle) override; + [[nodiscard]] bool IsTexturePixelFormatInteger(u32 handle) override; + [[nodiscard]] u32 ReadViewportTransformState() override; [[nodiscard]] u32 LocalMemorySize() const override; diff --git a/src/video_core/texture_cache/decode_bc.cpp b/src/video_core/texture_cache/decode_bc.cpp index 3e26474a3..a018c6df4 100644 --- a/src/video_core/texture_cache/decode_bc.cpp +++ b/src/video_core/texture_cache/decode_bc.cpp @@ -60,66 +60,72 @@ u32 ConvertedBytesPerBlock(VideoCore::Surface::PixelFormat pixel_format) { } template <auto decompress, PixelFormat pixel_format> -void DecompressBlocks(std::span<const u8> input, std::span<u8> output, Extent3D extent, +void DecompressBlocks(std::span<const u8> input, std::span<u8> output, BufferImageCopy& copy, bool is_signed = false) { const u32 out_bpp = ConvertedBytesPerBlock(pixel_format); - const u32 block_width = std::min(extent.width, BLOCK_SIZE); - const u32 block_height = std::min(extent.height, BLOCK_SIZE); - const u32 pitch = extent.width * out_bpp; + const u32 block_size = BlockSize(pixel_format); + const u32 width = copy.image_extent.width; + const u32 height = copy.image_extent.height * copy.image_subresource.num_layers; + const u32 depth = copy.image_extent.depth; + const u32 block_width = std::min(width, BLOCK_SIZE); + const u32 block_height = std::min(height, BLOCK_SIZE); + const u32 pitch = width * out_bpp; size_t input_offset = 0; size_t output_offset = 0; - for (u32 slice = 0; slice < extent.depth; ++slice) { - for (u32 y = 0; y < extent.height; y += block_height) { - size_t row_offset = 0; - for (u32 x = 0; x < extent.width; - x += block_width, row_offset += block_width * out_bpp) { - const u8* src = input.data() + input_offset; - u8* const dst = output.data() + output_offset + row_offset; + for (u32 slice = 0; slice < depth; ++slice) { + for (u32 y = 0; y < height; y += block_height) { + size_t src_offset = input_offset; + size_t dst_offset = output_offset; + for (u32 x = 0; x < width; x += block_width) { + const u8* src = input.data() + src_offset; + u8* const dst = output.data() + dst_offset; if constexpr (IsSigned(pixel_format)) { - decompress(src, dst, x, y, extent.width, extent.height, is_signed); + decompress(src, dst, x, y, width, height, is_signed); } else { - decompress(src, dst, x, y, extent.width, extent.height); + decompress(src, dst, x, y, width, height); } - input_offset += BlockSize(pixel_format); + src_offset += block_size; + dst_offset += block_width * out_bpp; } + input_offset += copy.buffer_row_length * block_size / block_width; output_offset += block_height * pitch; } } } -void DecompressBCn(std::span<const u8> input, std::span<u8> output, Extent3D extent, +void DecompressBCn(std::span<const u8> input, std::span<u8> output, BufferImageCopy& copy, VideoCore::Surface::PixelFormat pixel_format) { switch (pixel_format) { case PixelFormat::BC1_RGBA_UNORM: case PixelFormat::BC1_RGBA_SRGB: - DecompressBlocks<bcn::DecodeBc1, PixelFormat::BC1_RGBA_UNORM>(input, output, extent); + DecompressBlocks<bcn::DecodeBc1, PixelFormat::BC1_RGBA_UNORM>(input, output, copy); break; case PixelFormat::BC2_UNORM: case PixelFormat::BC2_SRGB: - DecompressBlocks<bcn::DecodeBc2, PixelFormat::BC2_UNORM>(input, output, extent); + DecompressBlocks<bcn::DecodeBc2, PixelFormat::BC2_UNORM>(input, output, copy); break; case PixelFormat::BC3_UNORM: case PixelFormat::BC3_SRGB: - DecompressBlocks<bcn::DecodeBc3, PixelFormat::BC3_UNORM>(input, output, extent); + DecompressBlocks<bcn::DecodeBc3, PixelFormat::BC3_UNORM>(input, output, copy); break; case PixelFormat::BC4_SNORM: case PixelFormat::BC4_UNORM: DecompressBlocks<bcn::DecodeBc4, PixelFormat::BC4_UNORM>( - input, output, extent, pixel_format == PixelFormat::BC4_SNORM); + input, output, copy, pixel_format == PixelFormat::BC4_SNORM); break; case PixelFormat::BC5_SNORM: case PixelFormat::BC5_UNORM: DecompressBlocks<bcn::DecodeBc5, PixelFormat::BC5_UNORM>( - input, output, extent, pixel_format == PixelFormat::BC5_SNORM); + input, output, copy, pixel_format == PixelFormat::BC5_SNORM); break; case PixelFormat::BC6H_SFLOAT: case PixelFormat::BC6H_UFLOAT: DecompressBlocks<bcn::DecodeBc6, PixelFormat::BC6H_UFLOAT>( - input, output, extent, pixel_format == PixelFormat::BC6H_SFLOAT); + input, output, copy, pixel_format == PixelFormat::BC6H_SFLOAT); break; case PixelFormat::BC7_SRGB: case PixelFormat::BC7_UNORM: - DecompressBlocks<bcn::DecodeBc7, PixelFormat::BC7_UNORM>(input, output, extent); + DecompressBlocks<bcn::DecodeBc7, PixelFormat::BC7_UNORM>(input, output, copy); break; default: LOG_WARNING(HW_GPU, "Unimplemented BCn decompression {}", pixel_format); diff --git a/src/video_core/texture_cache/decode_bc.h b/src/video_core/texture_cache/decode_bc.h index 41d1ec0a3..4e3b9b8ac 100644 --- a/src/video_core/texture_cache/decode_bc.h +++ b/src/video_core/texture_cache/decode_bc.h @@ -13,7 +13,7 @@ namespace VideoCommon { [[nodiscard]] u32 ConvertedBytesPerBlock(VideoCore::Surface::PixelFormat pixel_format); -void DecompressBCn(std::span<const u8> input, std::span<u8> output, Extent3D extent, +void DecompressBCn(std::span<const u8> input, std::span<u8> output, BufferImageCopy& copy, VideoCore::Surface::PixelFormat pixel_format); } // namespace VideoCommon diff --git a/src/video_core/texture_cache/util.cpp b/src/video_core/texture_cache/util.cpp index 15596c925..fcf70068e 100644 --- a/src/video_core/texture_cache/util.cpp +++ b/src/video_core/texture_cache/util.cpp @@ -837,6 +837,7 @@ boost::container::small_vector<BufferImageCopy, 16> UnswizzleImage(Tegra::Memory std::span<u8> output) { const size_t guest_size_bytes = input.size_bytes(); const u32 bpp_log2 = BytesPerBlockLog2(info.format); + const Extent2D tile_size = DefaultBlockSize(info.format); const Extent3D size = info.size; if (info.type == ImageType::Linear) { @@ -847,7 +848,7 @@ boost::container::small_vector<BufferImageCopy, 16> UnswizzleImage(Tegra::Memory return {{ .buffer_offset = 0, .buffer_size = guest_size_bytes, - .buffer_row_length = info.pitch >> bpp_log2, + .buffer_row_length = info.pitch * tile_size.width >> bpp_log2, .buffer_image_height = size.height, .image_subresource = { @@ -862,7 +863,6 @@ boost::container::small_vector<BufferImageCopy, 16> UnswizzleImage(Tegra::Memory const LevelInfo level_info = MakeLevelInfo(info); const s32 num_layers = info.resources.layers; const s32 num_levels = info.resources.levels; - const Extent2D tile_size = DefaultBlockSize(info.format); const std::array level_sizes = CalculateLevelSizes(level_info, num_levels); const Extent2D gob = GobSize(bpp_log2, info.block.height, info.tile_width_spacing); const u32 layer_size = CalculateLevelBytes(level_sizes, num_levels); @@ -926,8 +926,6 @@ void ConvertImage(std::span<const u8> input, const ImageInfo& info, std::span<u8 const auto input_offset = input.subspan(copy.buffer_offset); copy.buffer_offset = output_offset; - copy.buffer_row_length = mip_size.width; - copy.buffer_image_height = mip_size.height; const auto recompression_setting = Settings::values.astc_recompression.GetValue(); const bool astc = IsPixelFormatASTC(info.format); @@ -972,16 +970,14 @@ void ConvertImage(std::span<const u8> input, const ImageInfo& info, std::span<u8 bpp_div; output_offset += static_cast<u32>(copy.buffer_size); } else { - const Extent3D image_extent{ - .width = copy.image_extent.width, - .height = copy.image_extent.height * copy.image_subresource.num_layers, - .depth = copy.image_extent.depth, - }; - DecompressBCn(input_offset, output.subspan(output_offset), image_extent, info.format); + DecompressBCn(input_offset, output.subspan(output_offset), copy, info.format); output_offset += copy.image_extent.width * copy.image_extent.height * copy.image_subresource.num_layers * ConvertedBytesPerBlock(info.format); } + + copy.buffer_row_length = mip_size.width; + copy.buffer_image_height = mip_size.height; } } diff --git a/src/video_core/vulkan_common/vulkan_device.cpp b/src/video_core/vulkan_common/vulkan_device.cpp index 1fda0042d..727bbd98d 100644 --- a/src/video_core/vulkan_common/vulkan_device.cpp +++ b/src/video_core/vulkan_common/vulkan_device.cpp @@ -695,6 +695,11 @@ Device::Device(VkInstance instance_, vk::PhysicalDevice physical_, VkSurfaceKHR std::min(properties.properties.limits.maxVertexInputBindings, 16U); } + if (is_turnip) { + LOG_WARNING(Render_Vulkan, "Turnip requires higher-than-reported binding limits"); + properties.properties.limits.maxVertexInputBindings = 32; + } + if (!extensions.extended_dynamic_state && extensions.extended_dynamic_state2) { LOG_INFO(Render_Vulkan, "Removing extendedDynamicState2 due to missing extendedDynamicState"); @@ -750,10 +755,10 @@ VkFormat Device::GetSupportedFormat(VkFormat wanted_format, VkFormatFeatureFlags // The wanted format is not supported by hardware, search for alternatives const VkFormat* alternatives = GetFormatAlternatives(wanted_format); if (alternatives == nullptr) { - ASSERT_MSG(false, - "Format={} with usage={} and type={} has no defined alternatives and host " - "hardware does not support it", - wanted_format, wanted_usage, format_type); + LOG_ERROR(Render_Vulkan, + "Format={} with usage={} and type={} has no defined alternatives and host " + "hardware does not support it", + wanted_format, wanted_usage, format_type); return wanted_format; } @@ -769,10 +774,10 @@ VkFormat Device::GetSupportedFormat(VkFormat wanted_format, VkFormatFeatureFlags } // No alternatives found, panic - ASSERT_MSG(false, - "Format={} with usage={} and type={} is not supported by the host hardware and " - "doesn't support any of the alternatives", - wanted_format, wanted_usage, format_type); + LOG_ERROR(Render_Vulkan, + "Format={} with usage={} and type={} is not supported by the host hardware and " + "doesn't support any of the alternatives", + wanted_format, wanted_usage, format_type); return wanted_format; } diff --git a/src/video_core/vulkan_common/vulkan_device.h b/src/video_core/vulkan_common/vulkan_device.h index 4f3846345..701817086 100644 --- a/src/video_core/vulkan_common/vulkan_device.h +++ b/src/video_core/vulkan_common/vulkan_device.h @@ -665,6 +665,10 @@ public: return properties.properties.limits.maxViewports; } + u32 GetMaxUserClipDistances() const { + return properties.properties.limits.maxClipDistances; + } + bool SupportsConditionalBarriers() const { return supports_conditional_barriers; } diff --git a/src/video_core/vulkan_common/vulkan_wrapper.cpp b/src/video_core/vulkan_common/vulkan_wrapper.cpp index 70cf14afa..074aed964 100644 --- a/src/video_core/vulkan_common/vulkan_wrapper.cpp +++ b/src/video_core/vulkan_common/vulkan_wrapper.cpp @@ -246,7 +246,9 @@ void SetObjectName(const DeviceDispatch* dld, VkDevice device, T handle, VkObjec .objectHandle = reinterpret_cast<u64>(handle), .pObjectName = name, }; - Check(dld->vkSetDebugUtilsObjectNameEXT(device, &name_info)); + if (dld->vkSetDebugUtilsObjectNameEXT) { + Check(dld->vkSetDebugUtilsObjectNameEXT(device, &name_info)); + } } } // Anonymous namespace @@ -377,6 +379,8 @@ const char* ToString(VkResult result) noexcept { return "VK_OPERATION_DEFERRED_KHR"; case VkResult::VK_OPERATION_NOT_DEFERRED_KHR: return "VK_OPERATION_NOT_DEFERRED_KHR"; + case VkResult::VK_ERROR_INVALID_VIDEO_STD_PARAMETERS_KHR: + return "VK_ERROR_INVALID_VIDEO_STD_PARAMETERS_KHR"; case VkResult::VK_PIPELINE_COMPILE_REQUIRED_EXT: return "VK_PIPELINE_COMPILE_REQUIRED_EXT"; case VkResult::VK_RESULT_MAX_ENUM: |