From 420cc13248350ef5c2d19e0b961cb4185cd16a8a Mon Sep 17 00:00:00 2001 From: ReinUsesLisp Date: Sun, 17 May 2020 22:32:49 -0300 Subject: renderer_opengl: Add assembly program code paths Add code required to use OpenGL assembly programs based on NV_gpu_program5. Decompilation for ARB programs is intended to be added in a follow up commit. This does **not** include ARB decompilation and it's not in an usable state. The intention behind assembly programs is to reduce shader stutter significantly on drivers supporting NV_gpu_program5 (and other required extensions). Currently only Nvidia's proprietary driver supports these extensions. Add a UI option hidden for now to avoid people enabling this option accidentally. This code path has some limitations that OpenGL compatibility doesn't have: - NV_shader_storage_buffer_object is limited to 16 entries for a single OpenGL context state (I don't know if this is an intended limitation, an specification issue or I am missing something). Currently causes issues on The Legend of Zelda: Link's Awakening. - NV_parameter_buffer_object can't bind buffers using an offset different to zero. The used workaround is to copy to a temporary buffer (this doesn't happen often so it's not an issue). On the other hand, it has the following advantages: - Shaders build a lot faster. - We have control over how floating point rounding is done over individual instructions (SPIR-V on Vulkan can't do this). - Operations on shared memory can be unsigned and signed. - Transform feedbacks are dynamic state (not yet implemented). - Parameter buffers (uniform buffers) are per stage, matching NVN and hardware's behavior. - The API to bind and create assembly programs makes sense, unlike ARB_separate_shader_objects. --- src/video_core/renderer_opengl/gl_rasterizer.h | 16 +++++++++++----- 1 file changed, 11 insertions(+), 5 deletions(-) (limited to 'src/video_core/renderer_opengl/gl_rasterizer.h') diff --git a/src/video_core/renderer_opengl/gl_rasterizer.h b/src/video_core/renderer_opengl/gl_rasterizer.h index b94c65907..87f7fe159 100644 --- a/src/video_core/renderer_opengl/gl_rasterizer.h +++ b/src/video_core/renderer_opengl/gl_rasterizer.h @@ -56,8 +56,8 @@ struct DrawParameters; class RasterizerOpenGL : public VideoCore::RasterizerAccelerated { public: explicit RasterizerOpenGL(Core::System& system, Core::Frontend::EmuWindow& emu_window, - ScreenInfo& info, GLShader::ProgramManager& program_manager, - StateTracker& state_tracker); + const Device& device, ScreenInfo& info, + ProgramManager& program_manager, StateTracker& state_tracker); ~RasterizerOpenGL() override; void Draw(bool is_indexed, bool is_instanced) override; @@ -106,7 +106,7 @@ private: void SetupComputeConstBuffers(const Shader& kernel); /// Configures a constant buffer. - void SetupConstBuffer(u32 binding, const Tegra::Engines::ConstBufferInfo& buffer, + void SetupConstBuffer(GLenum stage, u32 binding, const Tegra::Engines::ConstBufferInfo& buffer, const ConstBufferEntry& entry); /// Configures the current global memory entries to use for the draw command. @@ -224,7 +224,7 @@ private: void SetupShaders(GLenum primitive_mode); - const Device device; + const Device& device; TextureCacheOpenGL texture_cache; ShaderCacheOpenGL shader_cache; @@ -236,7 +236,7 @@ private: Core::System& system; ScreenInfo& screen_info; - GLShader::ProgramManager& program_manager; + ProgramManager& program_manager; StateTracker& state_tracker; static constexpr std::size_t STREAM_BUFFER_SIZE = 128 * 1024 * 1024; @@ -248,6 +248,12 @@ private: std::bitset enabled_transform_feedback_buffers; + static constexpr std::size_t NUM_CONSTANT_BUFFERS = + Tegra::Engines::Maxwell3D::Regs::MaxConstBuffers * + Tegra::Engines::Maxwell3D::Regs::MaxShaderProgram; + std::array staging_cbufs{}; + std::size_t current_cbuf = 0; + /// Number of commands queued to the OpenGL driver. Reseted on flush. std::size_t num_queued_commands = 0; -- cgit v1.2.3