From 96ac3d518a9882a2a040f319c47a567467c9266d Mon Sep 17 00:00:00 2001 From: ReinUsesLisp Date: Wed, 25 Dec 2019 17:02:17 -0300 Subject: gl_rasterizer: Remove dirty flags --- src/video_core/engines/kepler_compute.cpp | 3 - src/video_core/engines/kepler_memory.cpp | 3 - src/video_core/engines/maxwell_3d.cpp | 181 +----------------------------- src/video_core/engines/maxwell_3d.h | 75 ------------- src/video_core/engines/maxwell_dma.cpp | 3 - 5 files changed, 1 insertion(+), 264 deletions(-) (limited to 'src/video_core/engines') diff --git a/src/video_core/engines/kepler_compute.cpp b/src/video_core/engines/kepler_compute.cpp index 4b824aa4e..254ad6810 100644 --- a/src/video_core/engines/kepler_compute.cpp +++ b/src/video_core/engines/kepler_compute.cpp @@ -38,9 +38,6 @@ void KeplerCompute::CallMethod(const GPU::MethodCall& method_call) { case KEPLER_COMPUTE_REG_INDEX(data_upload): { const bool is_last_call = method_call.IsLastCall(); upload_state.ProcessData(method_call.argument, is_last_call); - if (is_last_call) { - system.GPU().Maxwell3D().dirty.OnMemoryWrite(); - } break; } case KEPLER_COMPUTE_REG_INDEX(launch): diff --git a/src/video_core/engines/kepler_memory.cpp b/src/video_core/engines/kepler_memory.cpp index fa4a7c5c1..b504b450e 100644 --- a/src/video_core/engines/kepler_memory.cpp +++ b/src/video_core/engines/kepler_memory.cpp @@ -33,9 +33,6 @@ void KeplerMemory::CallMethod(const GPU::MethodCall& method_call) { case KEPLERMEMORY_REG_INDEX(data): { const bool is_last_call = method_call.IsLastCall(); upload_state.ProcessData(method_call.argument, is_last_call); - if (is_last_call) { - system.GPU().Maxwell3D().dirty.OnMemoryWrite(); - } break; } } diff --git a/src/video_core/engines/maxwell_3d.cpp b/src/video_core/engines/maxwell_3d.cpp index b28de1092..7a6bf764c 100644 --- a/src/video_core/engines/maxwell_3d.cpp +++ b/src/video_core/engines/maxwell_3d.cpp @@ -26,7 +26,6 @@ Maxwell3D::Maxwell3D(Core::System& system, VideoCore::RasterizerInterface& raste MemoryManager& memory_manager) : system{system}, rasterizer{rasterizer}, memory_manager{memory_manager}, macro_interpreter{*this}, upload_state{memory_manager, regs.upload} { - InitDirtySettings(); InitializeRegisterDefaults(); } @@ -103,164 +102,6 @@ void Maxwell3D::InitializeRegisterDefaults() { mme_inline[MAXWELL3D_REG_INDEX(index_array.count)] = true; } -#define DIRTY_REGS_POS(field_name) static_cast(offsetof(Maxwell3D::DirtyRegs, field_name)) - -void Maxwell3D::InitDirtySettings() { - const auto set_block = [this](std::size_t start, std::size_t range, u8 position) { - const auto start_itr = dirty_pointers.begin() + start; - const auto end_itr = start_itr + range; - std::fill(start_itr, end_itr, position); - }; - dirty.regs.fill(true); - - // Init Render Targets - constexpr u32 registers_per_rt = sizeof(regs.rt[0]) / sizeof(u32); - constexpr u32 rt_start_reg = MAXWELL3D_REG_INDEX(rt); - constexpr u32 rt_end_reg = rt_start_reg + registers_per_rt * 8; - u8 rt_dirty_reg = DIRTY_REGS_POS(render_target); - for (u32 rt_reg = rt_start_reg; rt_reg < rt_end_reg; rt_reg += registers_per_rt) { - set_block(rt_reg, registers_per_rt, rt_dirty_reg); - ++rt_dirty_reg; - } - constexpr u32 depth_buffer_flag = DIRTY_REGS_POS(depth_buffer); - dirty_pointers[MAXWELL3D_REG_INDEX(zeta_enable)] = depth_buffer_flag; - dirty_pointers[MAXWELL3D_REG_INDEX(zeta_width)] = depth_buffer_flag; - dirty_pointers[MAXWELL3D_REG_INDEX(zeta_height)] = depth_buffer_flag; - constexpr u32 registers_in_zeta = sizeof(regs.zeta) / sizeof(u32); - constexpr u32 zeta_reg = MAXWELL3D_REG_INDEX(zeta); - set_block(zeta_reg, registers_in_zeta, depth_buffer_flag); - - // Init Vertex Arrays - constexpr u32 vertex_array_start = MAXWELL3D_REG_INDEX(vertex_array); - constexpr u32 vertex_array_size = sizeof(regs.vertex_array[0]) / sizeof(u32); - constexpr u32 vertex_array_end = vertex_array_start + vertex_array_size * Regs::NumVertexArrays; - u8 va_dirty_reg = DIRTY_REGS_POS(vertex_array); - u8 vi_dirty_reg = DIRTY_REGS_POS(vertex_instance); - for (u32 vertex_reg = vertex_array_start; vertex_reg < vertex_array_end; - vertex_reg += vertex_array_size) { - set_block(vertex_reg, 3, va_dirty_reg); - // The divisor concerns vertex array instances - dirty_pointers[static_cast(vertex_reg) + 3] = vi_dirty_reg; - ++va_dirty_reg; - ++vi_dirty_reg; - } - constexpr u32 vertex_limit_start = MAXWELL3D_REG_INDEX(vertex_array_limit); - constexpr u32 vertex_limit_size = sizeof(regs.vertex_array_limit[0]) / sizeof(u32); - constexpr u32 vertex_limit_end = vertex_limit_start + vertex_limit_size * Regs::NumVertexArrays; - va_dirty_reg = DIRTY_REGS_POS(vertex_array); - for (u32 vertex_reg = vertex_limit_start; vertex_reg < vertex_limit_end; - vertex_reg += vertex_limit_size) { - set_block(vertex_reg, vertex_limit_size, va_dirty_reg); - va_dirty_reg++; - } - constexpr u32 vertex_instance_start = MAXWELL3D_REG_INDEX(instanced_arrays); - constexpr u32 vertex_instance_size = - sizeof(regs.instanced_arrays.is_instanced[0]) / sizeof(u32); - constexpr u32 vertex_instance_end = - vertex_instance_start + vertex_instance_size * Regs::NumVertexArrays; - vi_dirty_reg = DIRTY_REGS_POS(vertex_instance); - for (u32 vertex_reg = vertex_instance_start; vertex_reg < vertex_instance_end; - vertex_reg += vertex_instance_size) { - set_block(vertex_reg, vertex_instance_size, vi_dirty_reg); - vi_dirty_reg++; - } - set_block(MAXWELL3D_REG_INDEX(vertex_attrib_format), regs.vertex_attrib_format.size(), - DIRTY_REGS_POS(vertex_attrib_format)); - - // Init Shaders - constexpr u32 shader_registers_count = - sizeof(regs.shader_config[0]) * Regs::MaxShaderProgram / sizeof(u32); - set_block(MAXWELL3D_REG_INDEX(shader_config[0]), shader_registers_count, - DIRTY_REGS_POS(shaders)); - - // State - - // Viewport - constexpr u8 viewport_dirty_reg = DIRTY_REGS_POS(viewport); - constexpr u32 viewport_start = MAXWELL3D_REG_INDEX(viewports); - constexpr u32 viewport_size = sizeof(regs.viewports) / sizeof(u32); - set_block(viewport_start, viewport_size, viewport_dirty_reg); - constexpr u32 view_volume_start = MAXWELL3D_REG_INDEX(view_volume_clip_control); - constexpr u32 view_volume_size = sizeof(regs.view_volume_clip_control) / sizeof(u32); - set_block(view_volume_start, view_volume_size, viewport_dirty_reg); - - // Viewport transformation - constexpr u32 viewport_trans_start = MAXWELL3D_REG_INDEX(viewport_transform); - constexpr u32 viewport_trans_size = sizeof(regs.viewport_transform) / sizeof(u32); - set_block(viewport_trans_start, viewport_trans_size, DIRTY_REGS_POS(viewport_transform)); - - // Cullmode - constexpr u32 cull_mode_start = MAXWELL3D_REG_INDEX(cull); - constexpr u32 cull_mode_size = sizeof(regs.cull) / sizeof(u32); - set_block(cull_mode_start, cull_mode_size, DIRTY_REGS_POS(cull_mode)); - - // Screen y control - dirty_pointers[MAXWELL3D_REG_INDEX(screen_y_control)] = DIRTY_REGS_POS(screen_y_control); - - // Primitive Restart - constexpr u32 primitive_restart_start = MAXWELL3D_REG_INDEX(primitive_restart); - constexpr u32 primitive_restart_size = sizeof(regs.primitive_restart) / sizeof(u32); - set_block(primitive_restart_start, primitive_restart_size, DIRTY_REGS_POS(primitive_restart)); - - // Depth Test - constexpr u8 depth_test_dirty_reg = DIRTY_REGS_POS(depth_test); - dirty_pointers[MAXWELL3D_REG_INDEX(depth_test_enable)] = depth_test_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(depth_write_enabled)] = depth_test_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(depth_test_func)] = depth_test_dirty_reg; - - // Stencil Test - constexpr u32 stencil_test_dirty_reg = DIRTY_REGS_POS(stencil_test); - dirty_pointers[MAXWELL3D_REG_INDEX(stencil_enable)] = stencil_test_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(stencil_front_func_func)] = stencil_test_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(stencil_front_func_ref)] = stencil_test_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(stencil_front_func_mask)] = stencil_test_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(stencil_front_op_fail)] = stencil_test_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(stencil_front_op_zfail)] = stencil_test_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(stencil_front_op_zpass)] = stencil_test_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(stencil_front_mask)] = stencil_test_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(stencil_two_side_enable)] = stencil_test_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(stencil_back_func_func)] = stencil_test_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(stencil_back_func_ref)] = stencil_test_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(stencil_back_func_mask)] = stencil_test_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(stencil_back_op_fail)] = stencil_test_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(stencil_back_op_zfail)] = stencil_test_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(stencil_back_op_zpass)] = stencil_test_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(stencil_back_mask)] = stencil_test_dirty_reg; - - // Color Mask - constexpr u8 color_mask_dirty_reg = DIRTY_REGS_POS(color_mask); - dirty_pointers[MAXWELL3D_REG_INDEX(color_mask_common)] = color_mask_dirty_reg; - set_block(MAXWELL3D_REG_INDEX(color_mask), sizeof(regs.color_mask) / sizeof(u32), - color_mask_dirty_reg); - // Blend State - constexpr u8 blend_state_dirty_reg = DIRTY_REGS_POS(blend_state); - set_block(MAXWELL3D_REG_INDEX(blend_color), sizeof(regs.blend_color) / sizeof(u32), - blend_state_dirty_reg); - dirty_pointers[MAXWELL3D_REG_INDEX(independent_blend_enable)] = blend_state_dirty_reg; - set_block(MAXWELL3D_REG_INDEX(blend), sizeof(regs.blend) / sizeof(u32), blend_state_dirty_reg); - set_block(MAXWELL3D_REG_INDEX(independent_blend), sizeof(regs.independent_blend) / sizeof(u32), - blend_state_dirty_reg); - - // Scissor State - constexpr u8 scissor_test_dirty_reg = DIRTY_REGS_POS(scissor_test); - set_block(MAXWELL3D_REG_INDEX(scissor_test), sizeof(regs.scissor_test) / sizeof(u32), - scissor_test_dirty_reg); - - // Polygon Offset - constexpr u8 polygon_offset_dirty_reg = DIRTY_REGS_POS(polygon_offset); - dirty_pointers[MAXWELL3D_REG_INDEX(polygon_offset_fill_enable)] = polygon_offset_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(polygon_offset_line_enable)] = polygon_offset_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(polygon_offset_point_enable)] = polygon_offset_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(polygon_offset_units)] = polygon_offset_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(polygon_offset_factor)] = polygon_offset_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(polygon_offset_clamp)] = polygon_offset_dirty_reg; - - // Depth bounds - constexpr u8 depth_bounds_values_dirty_reg = DIRTY_REGS_POS(depth_bounds_values); - dirty_pointers[MAXWELL3D_REG_INDEX(depth_bounds[0])] = depth_bounds_values_dirty_reg; - dirty_pointers[MAXWELL3D_REG_INDEX(depth_bounds[1])] = depth_bounds_values_dirty_reg; -} - void Maxwell3D::CallMacroMethod(u32 method, std::size_t num_parameters, const u32* parameters) { // Reset the current macro. executing_macro = 0; @@ -317,23 +158,7 @@ void Maxwell3D::CallMethod(const GPU::MethodCall& method_call) { ASSERT_MSG(method < Regs::NUM_REGS, "Invalid Maxwell3D register, increase the size of the Regs structure"); - if (regs.reg_array[method] != method_call.argument) { - regs.reg_array[method] = method_call.argument; - const std::size_t dirty_reg = dirty_pointers[method]; - if (dirty_reg) { - dirty.regs[dirty_reg] = true; - if (dirty_reg >= DIRTY_REGS_POS(vertex_array) && - dirty_reg < DIRTY_REGS_POS(vertex_array_buffers)) { - dirty.vertex_array_buffers = true; - } else if (dirty_reg >= DIRTY_REGS_POS(vertex_instance) && - dirty_reg < DIRTY_REGS_POS(vertex_instances)) { - dirty.vertex_instances = true; - } else if (dirty_reg >= DIRTY_REGS_POS(render_target) && - dirty_reg < DIRTY_REGS_POS(render_settings)) { - dirty.render_settings = true; - } - } - } + regs.reg_array[method] = method_call.argument; switch (method) { case MAXWELL3D_REG_INDEX(macros.data): { @@ -418,9 +243,6 @@ void Maxwell3D::CallMethod(const GPU::MethodCall& method_call) { case MAXWELL3D_REG_INDEX(data_upload): { const bool is_last_call = method_call.IsLastCall(); upload_state.ProcessData(method_call.argument, is_last_call); - if (is_last_call) { - dirty.OnMemoryWrite(); - } break; } default: @@ -727,7 +549,6 @@ void Maxwell3D::FinishCBData() { const u32 id = cb_data_state.id; memory_manager.WriteBlock(address, cb_data_state.buffer[id].data(), size); - dirty.OnMemoryWrite(); cb_data_state.id = null_cb_data; cb_data_state.current = null_cb_data; diff --git a/src/video_core/engines/maxwell_3d.h b/src/video_core/engines/maxwell_3d.h index 6ea7cc6a5..3a641c182 100644 --- a/src/video_core/engines/maxwell_3d.h +++ b/src/video_core/engines/maxwell_3d.h @@ -1238,79 +1238,6 @@ public: State state{}; - struct DirtyRegs { - static constexpr std::size_t NUM_REGS = 256; - static_assert(NUM_REGS - 1 <= std::numeric_limits::max()); - - union { - struct { - bool null_dirty; - - // Vertex Attributes - bool vertex_attrib_format; - - // Vertex Arrays - std::array vertex_array; - - bool vertex_array_buffers; - - // Vertex Instances - std::array vertex_instance; - - bool vertex_instances; - - // Render Targets - std::array render_target; - bool depth_buffer; - - bool render_settings; - - // Shaders - bool shaders; - - // Rasterizer State - bool viewport; - bool clip_coefficient; - bool cull_mode; - bool primitive_restart; - bool depth_test; - bool stencil_test; - bool blend_state; - bool scissor_test; - bool transform_feedback; - bool color_mask; - bool polygon_offset; - bool depth_bounds_values; - - // Complementary - bool viewport_transform; - bool screen_y_control; - - bool memory_general; - }; - std::array regs; - }; - - void ResetVertexArrays() { - vertex_array.fill(true); - vertex_array_buffers = true; - } - - void ResetRenderTargets() { - depth_buffer = true; - render_target.fill(true); - render_settings = true; - } - - void OnMemoryWrite() { - shaders = true; - memory_general = true; - ResetRenderTargets(); - ResetVertexArrays(); - } - - } dirty{}; - /// Reads a register value located at the input method address u32 GetRegisterValue(u32 method) const; @@ -1417,8 +1344,6 @@ private: /// Retrieves information about a specific TSC entry from the TSC buffer. Texture::TSCEntry GetTSCEntry(u32 tsc_index) const; - void InitDirtySettings(); - /** * Call a macro on this engine. * @param method Method to call diff --git a/src/video_core/engines/maxwell_dma.cpp b/src/video_core/engines/maxwell_dma.cpp index ad8453c5f..ae51765a6 100644 --- a/src/video_core/engines/maxwell_dma.cpp +++ b/src/video_core/engines/maxwell_dma.cpp @@ -56,9 +56,6 @@ void MaxwellDMA::HandleCopy() { return; } - // All copies here update the main memory, so mark all rasterizer states as invalid. - system.GPU().Maxwell3D().dirty.OnMemoryWrite(); - if (regs.exec.is_dst_linear && regs.exec.is_src_linear) { // When the enable_2d bit is disabled, the copy is performed as if we were copying a 1D // buffer of length `x_count`, otherwise we copy a 2D image of dimensions (x_count, -- cgit v1.2.3 From d3e433a38048c5d32c0929446008586e975ccd0e Mon Sep 17 00:00:00 2001 From: ReinUsesLisp Date: Thu, 26 Dec 2019 01:50:38 -0300 Subject: gl_state: Remove viewport and depth range tracking --- src/video_core/engines/maxwell_3d.h | 18 +++++++++--------- 1 file changed, 9 insertions(+), 9 deletions(-) (limited to 'src/video_core/engines') diff --git a/src/video_core/engines/maxwell_3d.h b/src/video_core/engines/maxwell_3d.h index 3a641c182..2134d6e4f 100644 --- a/src/video_core/engines/maxwell_3d.h +++ b/src/video_core/engines/maxwell_3d.h @@ -574,7 +574,7 @@ public: f32 translate_z; INSERT_UNION_PADDING_WORDS(2); - Common::Rectangle GetRect() const { + Common::Rectangle GetRect() const { return { GetX(), // left GetY() + GetHeight(), // top @@ -583,20 +583,20 @@ public: }; }; - s32 GetX() const { - return static_cast(std::max(0.0f, translate_x - std::fabs(scale_x))); + f32 GetX() const { + return std::max(0.0f, translate_x - std::fabs(scale_x)); } - s32 GetY() const { - return static_cast(std::max(0.0f, translate_y - std::fabs(scale_y))); + f32 GetY() const { + return std::max(0.0f, translate_y - std::fabs(scale_y)); } - s32 GetWidth() const { - return static_cast(translate_x + std::fabs(scale_x)) - GetX(); + f32 GetWidth() const { + return translate_x + std::fabs(scale_x) - GetX(); } - s32 GetHeight() const { - return static_cast(translate_y + std::fabs(scale_y)) - GetY(); + f32 GetHeight() const { + return translate_y + std::fabs(scale_y) - GetY(); } }; -- cgit v1.2.3 From 1eee891f6e73423a9aa6147f980be5aea799e7ce Mon Sep 17 00:00:00 2001 From: ReinUsesLisp Date: Thu, 26 Dec 2019 02:20:08 -0300 Subject: gl_state: Remove clip distances tracking --- src/video_core/engines/maxwell_3d.h | 11 +---------- 1 file changed, 1 insertion(+), 10 deletions(-) (limited to 'src/video_core/engines') diff --git a/src/video_core/engines/maxwell_3d.h b/src/video_core/engines/maxwell_3d.h index 2134d6e4f..b0fb0fb7d 100644 --- a/src/video_core/engines/maxwell_3d.h +++ b/src/video_core/engines/maxwell_3d.h @@ -872,16 +872,7 @@ public: INSERT_UNION_PADDING_WORDS(0x35); - union { - BitField<0, 1, u32> c0; - BitField<1, 1, u32> c1; - BitField<2, 1, u32> c2; - BitField<3, 1, u32> c3; - BitField<4, 1, u32> c4; - BitField<5, 1, u32> c5; - BitField<6, 1, u32> c6; - BitField<7, 1, u32> c7; - } clip_distance_enabled; + u32 clip_distance_enabled; u32 samplecnt_enable; -- cgit v1.2.3 From eed789d0d134fbeef1c16f9829b5c1b4b7dabb17 Mon Sep 17 00:00:00 2001 From: ReinUsesLisp Date: Thu, 26 Dec 2019 22:14:10 -0300 Subject: video_core: Reintroduce dirty flags infrastructure --- src/video_core/engines/kepler_compute.cpp | 3 +++ src/video_core/engines/kepler_memory.cpp | 3 +++ src/video_core/engines/maxwell_3d.cpp | 14 +++++++++++++- src/video_core/engines/maxwell_3d.h | 14 ++++++++++++++ src/video_core/engines/maxwell_dma.cpp | 3 +++ 5 files changed, 36 insertions(+), 1 deletion(-) (limited to 'src/video_core/engines') diff --git a/src/video_core/engines/kepler_compute.cpp b/src/video_core/engines/kepler_compute.cpp index 254ad6810..ae52afa79 100644 --- a/src/video_core/engines/kepler_compute.cpp +++ b/src/video_core/engines/kepler_compute.cpp @@ -38,6 +38,9 @@ void KeplerCompute::CallMethod(const GPU::MethodCall& method_call) { case KEPLER_COMPUTE_REG_INDEX(data_upload): { const bool is_last_call = method_call.IsLastCall(); upload_state.ProcessData(method_call.argument, is_last_call); + if (is_last_call) { + system.GPU().Maxwell3D().OnMemoryWrite(); + } break; } case KEPLER_COMPUTE_REG_INDEX(launch): diff --git a/src/video_core/engines/kepler_memory.cpp b/src/video_core/engines/kepler_memory.cpp index b504b450e..597872e43 100644 --- a/src/video_core/engines/kepler_memory.cpp +++ b/src/video_core/engines/kepler_memory.cpp @@ -33,6 +33,9 @@ void KeplerMemory::CallMethod(const GPU::MethodCall& method_call) { case KEPLERMEMORY_REG_INDEX(data): { const bool is_last_call = method_call.IsLastCall(); upload_state.ProcessData(method_call.argument, is_last_call); + if (is_last_call) { + system.GPU().Maxwell3D().OnMemoryWrite(); + } break; } } diff --git a/src/video_core/engines/maxwell_3d.cpp b/src/video_core/engines/maxwell_3d.cpp index 7a6bf764c..db710bf35 100644 --- a/src/video_core/engines/maxwell_3d.cpp +++ b/src/video_core/engines/maxwell_3d.cpp @@ -26,6 +26,8 @@ Maxwell3D::Maxwell3D(Core::System& system, VideoCore::RasterizerInterface& raste MemoryManager& memory_manager) : system{system}, rasterizer{rasterizer}, memory_manager{memory_manager}, macro_interpreter{*this}, upload_state{memory_manager, regs.upload} { + dirty.flags.flip(); + InitializeRegisterDefaults(); } @@ -158,7 +160,13 @@ void Maxwell3D::CallMethod(const GPU::MethodCall& method_call) { ASSERT_MSG(method < Regs::NUM_REGS, "Invalid Maxwell3D register, increase the size of the Regs structure"); - regs.reg_array[method] = method_call.argument; + if (regs.reg_array[method] != method_call.argument) { + regs.reg_array[method] = method_call.argument; + + for (const auto& table : dirty.tables) { + dirty.flags[table[method]] = true; + } + } switch (method) { case MAXWELL3D_REG_INDEX(macros.data): { @@ -243,6 +251,9 @@ void Maxwell3D::CallMethod(const GPU::MethodCall& method_call) { case MAXWELL3D_REG_INDEX(data_upload): { const bool is_last_call = method_call.IsLastCall(); upload_state.ProcessData(method_call.argument, is_last_call); + if (is_last_call) { + OnMemoryWrite(); + } break; } default: @@ -549,6 +560,7 @@ void Maxwell3D::FinishCBData() { const u32 id = cb_data_state.id; memory_manager.WriteBlock(address, cb_data_state.buffer[id].data(), size); + OnMemoryWrite(); cb_data_state.id = null_cb_data; cb_data_state.current = null_cb_data; diff --git a/src/video_core/engines/maxwell_3d.h b/src/video_core/engines/maxwell_3d.h index b0fb0fb7d..72848b1e8 100644 --- a/src/video_core/engines/maxwell_3d.h +++ b/src/video_core/engines/maxwell_3d.h @@ -6,6 +6,7 @@ #include #include +#include #include #include #include @@ -1274,6 +1275,13 @@ public: return execute_on; } + /// Notify a memory write has happened. + void OnMemoryWrite() { + for (const u8 store : dirty.on_write_stores) { + dirty.flags[store] = true; + } + } + enum class MMEDrawMode : u32 { Undefined, Array, @@ -1289,6 +1297,12 @@ public: u32 gl_end_count{}; } mme_draw; + struct { + std::bitset::max()> flags; + std::array, 3> tables{}; + std::array on_write_stores{}; + } dirty; + private: void InitializeRegisterDefaults(); diff --git a/src/video_core/engines/maxwell_dma.cpp b/src/video_core/engines/maxwell_dma.cpp index ae51765a6..c2610f992 100644 --- a/src/video_core/engines/maxwell_dma.cpp +++ b/src/video_core/engines/maxwell_dma.cpp @@ -56,6 +56,9 @@ void MaxwellDMA::HandleCopy() { return; } + // All copies here update the main memory, so mark all rasterizer states as invalid. + system.GPU().Maxwell3D().OnMemoryWrite(); + if (regs.exec.is_dst_linear && regs.exec.is_src_linear) { // When the enable_2d bit is disabled, the copy is performed as if we were copying a 1D // buffer of length `x_count`, otherwise we copy a 2D image of dimensions (x_count, -- cgit v1.2.3 From 9e74e6988b881c6889074bd2335239eb2e491e91 Mon Sep 17 00:00:00 2001 From: ReinUsesLisp Date: Sat, 28 Dec 2019 21:41:41 -0300 Subject: maxwell_3d: Flatten cull and front face registers --- src/video_core/engines/maxwell_3d.cpp | 6 +++--- src/video_core/engines/maxwell_3d.h | 30 ++++++++++++++---------------- 2 files changed, 17 insertions(+), 19 deletions(-) (limited to 'src/video_core/engines') diff --git a/src/video_core/engines/maxwell_3d.cpp b/src/video_core/engines/maxwell_3d.cpp index db710bf35..89050361e 100644 --- a/src/video_core/engines/maxwell_3d.cpp +++ b/src/video_core/engines/maxwell_3d.cpp @@ -76,8 +76,8 @@ void Maxwell3D::InitializeRegisterDefaults() { regs.stencil_back_mask = 0xFFFFFFFF; regs.depth_test_func = Regs::ComparisonOp::Always; - regs.cull.front_face = Regs::Cull::FrontFace::CounterClockWise; - regs.cull.cull_face = Regs::Cull::CullFace::Back; + regs.front_face = Regs::FrontFace::CounterClockWise; + regs.cull_face = Regs::CullFace::Back; // TODO(Rodrigo): Most games do not set a point size. I think this is a case of a // register carrying a default value. Assume it's OpenGL's default (1). @@ -96,7 +96,7 @@ void Maxwell3D::InitializeRegisterDefaults() { regs.rasterize_enable = 1; regs.rt_separate_frag_data = 1; regs.framebuffer_srgb = 1; - regs.cull.front_face = Maxwell3D::Regs::Cull::FrontFace::ClockWise; + regs.front_face = Maxwell3D::Regs::FrontFace::ClockWise; mme_inline[MAXWELL3D_REG_INDEX(draw.vertex_end_gl)] = true; mme_inline[MAXWELL3D_REG_INDEX(draw.vertex_begin_gl)] = true; diff --git a/src/video_core/engines/maxwell_3d.h b/src/video_core/engines/maxwell_3d.h index 72848b1e8..8edfa6a34 100644 --- a/src/video_core/engines/maxwell_3d.h +++ b/src/video_core/engines/maxwell_3d.h @@ -432,21 +432,15 @@ public: GeneratedPrimitives = 0x1F, }; - struct Cull { - enum class FrontFace : u32 { - ClockWise = 0x0900, - CounterClockWise = 0x0901, - }; - - enum class CullFace : u32 { - Front = 0x0404, - Back = 0x0405, - FrontAndBack = 0x0408, - }; + enum class FrontFace : u32 { + ClockWise = 0x0900, + CounterClockWise = 0x0901, + }; - u32 enabled; - FrontFace front_face; - CullFace cull_face; + enum class CullFace : u32 { + Front = 0x0404, + Back = 0x0405, + FrontAndBack = 0x0408, }; struct Blend { @@ -1052,7 +1046,9 @@ public: INSERT_UNION_PADDING_WORDS(1); - Cull cull; + u32 cull_test_enabled; + FrontFace front_face; + CullFace cull_face; u32 pixel_center_integer; @@ -1491,7 +1487,9 @@ ASSERT_REG_POSITION(index_array, 0x5F2); ASSERT_REG_POSITION(polygon_offset_clamp, 0x61F); ASSERT_REG_POSITION(instanced_arrays, 0x620); ASSERT_REG_POSITION(vp_point_size, 0x644); -ASSERT_REG_POSITION(cull, 0x646); +ASSERT_REG_POSITION(cull_test_enabled, 0x646); +ASSERT_REG_POSITION(front_face, 0x647); +ASSERT_REG_POSITION(cull_face, 0x648); ASSERT_REG_POSITION(pixel_center_integer, 0x649); ASSERT_REG_POSITION(viewport_transform_enabled, 0x64B); ASSERT_REG_POSITION(view_volume_clip_control, 0x64F); -- cgit v1.2.3 From 9b08698a0cd1c958a4479ca544dc35333aa0e370 Mon Sep 17 00:00:00 2001 From: ReinUsesLisp Date: Sun, 29 Dec 2019 01:05:02 -0300 Subject: maxwell_3d: Change write dirty flags to a bitset --- src/video_core/engines/maxwell_3d.h | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'src/video_core/engines') diff --git a/src/video_core/engines/maxwell_3d.h b/src/video_core/engines/maxwell_3d.h index 8edfa6a34..beaf3ffb6 100644 --- a/src/video_core/engines/maxwell_3d.h +++ b/src/video_core/engines/maxwell_3d.h @@ -1273,9 +1273,7 @@ public: /// Notify a memory write has happened. void OnMemoryWrite() { - for (const u8 store : dirty.on_write_stores) { - dirty.flags[store] = true; - } + dirty.flags |= dirty.on_write_stores; } enum class MMEDrawMode : u32 { @@ -1295,8 +1293,8 @@ public: struct { std::bitset::max()> flags; + std::bitset::max()> on_write_stores; std::array, 3> tables{}; - std::array on_write_stores{}; } dirty; private: -- cgit v1.2.3 From 15cadc394848619be5e0ccaa43dc00488476218d Mon Sep 17 00:00:00 2001 From: ReinUsesLisp Date: Thu, 2 Jan 2020 22:27:52 -0300 Subject: maxwell_3d: Use two tables instead of three for dirty flags --- src/video_core/engines/maxwell_3d.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'src/video_core/engines') diff --git a/src/video_core/engines/maxwell_3d.h b/src/video_core/engines/maxwell_3d.h index beaf3ffb6..3ff6dec75 100644 --- a/src/video_core/engines/maxwell_3d.h +++ b/src/video_core/engines/maxwell_3d.h @@ -1294,7 +1294,7 @@ public: struct { std::bitset::max()> flags; std::bitset::max()> on_write_stores; - std::array, 3> tables{}; + std::array, 2> tables{}; } dirty; private: -- cgit v1.2.3 From 042256c6bbbe27a71805aa2dabe2cac436134b3d Mon Sep 17 00:00:00 2001 From: ReinUsesLisp Date: Fri, 21 Feb 2020 01:19:07 -0300 Subject: state_tracker: Remove type traits with named structures --- src/video_core/engines/maxwell_3d.h | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-) (limited to 'src/video_core/engines') diff --git a/src/video_core/engines/maxwell_3d.h b/src/video_core/engines/maxwell_3d.h index 3ff6dec75..491cff370 100644 --- a/src/video_core/engines/maxwell_3d.h +++ b/src/video_core/engines/maxwell_3d.h @@ -1291,10 +1291,14 @@ public: u32 gl_end_count{}; } mme_draw; - struct { - std::bitset::max()> flags; - std::bitset::max()> on_write_stores; - std::array, 2> tables{}; + struct DirtyState { + using Flags = std::bitset::max()>; + using Table = std::array; + using Tables = std::array; + + Flags flags; + Flags on_write_stores; + Tables tables{}; } dirty; private: -- cgit v1.2.3