gl_rasterizer: Remove dirty flags

This commit is contained in:
ReinUsesLisp 2019-12-25 17:02:17 -03:00
parent e38ed26b98
commit 96ac3d518a
18 changed files with 7 additions and 457 deletions

View file

@ -69,6 +69,8 @@ add_library(video_core STATIC
renderer_opengl/gl_shader_manager.h renderer_opengl/gl_shader_manager.h
renderer_opengl/gl_shader_util.cpp renderer_opengl/gl_shader_util.cpp
renderer_opengl/gl_shader_util.h renderer_opengl/gl_shader_util.h
renderer_opengl/gl_state_tracker.cpp
renderer_opengl/gl_state_tracker.h
renderer_opengl/gl_state.cpp renderer_opengl/gl_state.cpp
renderer_opengl/gl_state.h renderer_opengl/gl_state.h
renderer_opengl/gl_stream_buffer.cpp renderer_opengl/gl_stream_buffer.cpp

View file

@ -21,9 +21,6 @@ MICROPROFILE_DEFINE(DispatchCalls, "GPU", "Execute command buffer", MP_RGB(128,
void DmaPusher::DispatchCalls() { void DmaPusher::DispatchCalls() {
MICROPROFILE_SCOPE(DispatchCalls); MICROPROFILE_SCOPE(DispatchCalls);
// On entering GPU code, assume all memory may be touched by the ARM core.
gpu.Maxwell3D().dirty.OnMemoryWrite();
dma_pushbuffer_subindex = 0; dma_pushbuffer_subindex = 0;
while (Core::System::GetInstance().IsPoweredOn()) { while (Core::System::GetInstance().IsPoweredOn()) {

View file

@ -38,9 +38,6 @@ void KeplerCompute::CallMethod(const GPU::MethodCall& method_call) {
case KEPLER_COMPUTE_REG_INDEX(data_upload): { case KEPLER_COMPUTE_REG_INDEX(data_upload): {
const bool is_last_call = method_call.IsLastCall(); const bool is_last_call = method_call.IsLastCall();
upload_state.ProcessData(method_call.argument, is_last_call); upload_state.ProcessData(method_call.argument, is_last_call);
if (is_last_call) {
system.GPU().Maxwell3D().dirty.OnMemoryWrite();
}
break; break;
} }
case KEPLER_COMPUTE_REG_INDEX(launch): case KEPLER_COMPUTE_REG_INDEX(launch):

View file

@ -33,9 +33,6 @@ void KeplerMemory::CallMethod(const GPU::MethodCall& method_call) {
case KEPLERMEMORY_REG_INDEX(data): { case KEPLERMEMORY_REG_INDEX(data): {
const bool is_last_call = method_call.IsLastCall(); const bool is_last_call = method_call.IsLastCall();
upload_state.ProcessData(method_call.argument, is_last_call); upload_state.ProcessData(method_call.argument, is_last_call);
if (is_last_call) {
system.GPU().Maxwell3D().dirty.OnMemoryWrite();
}
break; break;
} }
} }

View file

@ -26,7 +26,6 @@ Maxwell3D::Maxwell3D(Core::System& system, VideoCore::RasterizerInterface& raste
MemoryManager& memory_manager) MemoryManager& memory_manager)
: system{system}, rasterizer{rasterizer}, memory_manager{memory_manager}, : system{system}, rasterizer{rasterizer}, memory_manager{memory_manager},
macro_interpreter{*this}, upload_state{memory_manager, regs.upload} { macro_interpreter{*this}, upload_state{memory_manager, regs.upload} {
InitDirtySettings();
InitializeRegisterDefaults(); InitializeRegisterDefaults();
} }
@ -103,164 +102,6 @@ void Maxwell3D::InitializeRegisterDefaults() {
mme_inline[MAXWELL3D_REG_INDEX(index_array.count)] = true; mme_inline[MAXWELL3D_REG_INDEX(index_array.count)] = true;
} }
#define DIRTY_REGS_POS(field_name) static_cast<u8>(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<std::size_t>(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) { void Maxwell3D::CallMacroMethod(u32 method, std::size_t num_parameters, const u32* parameters) {
// Reset the current macro. // Reset the current macro.
executing_macro = 0; executing_macro = 0;
@ -317,23 +158,7 @@ void Maxwell3D::CallMethod(const GPU::MethodCall& method_call) {
ASSERT_MSG(method < Regs::NUM_REGS, ASSERT_MSG(method < Regs::NUM_REGS,
"Invalid Maxwell3D register, increase the size of the Regs structure"); "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; 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;
}
}
}
switch (method) { switch (method) {
case MAXWELL3D_REG_INDEX(macros.data): { case MAXWELL3D_REG_INDEX(macros.data): {
@ -418,9 +243,6 @@ void Maxwell3D::CallMethod(const GPU::MethodCall& method_call) {
case MAXWELL3D_REG_INDEX(data_upload): { case MAXWELL3D_REG_INDEX(data_upload): {
const bool is_last_call = method_call.IsLastCall(); const bool is_last_call = method_call.IsLastCall();
upload_state.ProcessData(method_call.argument, is_last_call); upload_state.ProcessData(method_call.argument, is_last_call);
if (is_last_call) {
dirty.OnMemoryWrite();
}
break; break;
} }
default: default:
@ -727,7 +549,6 @@ void Maxwell3D::FinishCBData() {
const u32 id = cb_data_state.id; const u32 id = cb_data_state.id;
memory_manager.WriteBlock(address, cb_data_state.buffer[id].data(), size); memory_manager.WriteBlock(address, cb_data_state.buffer[id].data(), size);
dirty.OnMemoryWrite();
cb_data_state.id = null_cb_data; cb_data_state.id = null_cb_data;
cb_data_state.current = null_cb_data; cb_data_state.current = null_cb_data;

View file

@ -1238,79 +1238,6 @@ public:
State state{}; State state{};
struct DirtyRegs {
static constexpr std::size_t NUM_REGS = 256;
static_assert(NUM_REGS - 1 <= std::numeric_limits<u8>::max());
union {
struct {
bool null_dirty;
// Vertex Attributes
bool vertex_attrib_format;
// Vertex Arrays
std::array<bool, 32> vertex_array;
bool vertex_array_buffers;
// Vertex Instances
std::array<bool, 32> vertex_instance;
bool vertex_instances;
// Render Targets
std::array<bool, 8> 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<bool, NUM_REGS> 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 /// Reads a register value located at the input method address
u32 GetRegisterValue(u32 method) const; u32 GetRegisterValue(u32 method) const;
@ -1417,8 +1344,6 @@ private:
/// Retrieves information about a specific TSC entry from the TSC buffer. /// Retrieves information about a specific TSC entry from the TSC buffer.
Texture::TSCEntry GetTSCEntry(u32 tsc_index) const; Texture::TSCEntry GetTSCEntry(u32 tsc_index) const;
void InitDirtySettings();
/** /**
* Call a macro on this engine. * Call a macro on this engine.
* @param method Method to call * @param method Method to call

View file

@ -56,9 +56,6 @@ void MaxwellDMA::HandleCopy() {
return; 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) { 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 // 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, // buffer of length `x_count`, otherwise we copy a 2D image of dimensions (x_count,

View file

@ -117,11 +117,6 @@ GLuint RasterizerOpenGL::SetupVertexFormat() {
auto& gpu = system.GPU().Maxwell3D(); auto& gpu = system.GPU().Maxwell3D();
const auto& regs = gpu.regs; const auto& regs = gpu.regs;
if (!gpu.dirty.vertex_attrib_format) {
return state.draw.vertex_array;
}
gpu.dirty.vertex_attrib_format = false;
MICROPROFILE_SCOPE(OpenGL_VAO); MICROPROFILE_SCOPE(OpenGL_VAO);
auto [iter, is_cache_miss] = vertex_array_cache.try_emplace(regs.vertex_attrib_format); auto [iter, is_cache_miss] = vertex_array_cache.try_emplace(regs.vertex_attrib_format);
@ -173,30 +168,18 @@ GLuint RasterizerOpenGL::SetupVertexFormat() {
} }
} }
// Rebinding the VAO invalidates the vertex buffer bindings.
gpu.dirty.ResetVertexArrays();
state.draw.vertex_array = vao_entry.handle; state.draw.vertex_array = vao_entry.handle;
return vao_entry.handle; return vao_entry.handle;
} }
void RasterizerOpenGL::SetupVertexBuffer(GLuint vao) { void RasterizerOpenGL::SetupVertexBuffer(GLuint vao) {
auto& gpu = system.GPU().Maxwell3D(); auto& gpu = system.GPU().Maxwell3D();
if (!gpu.dirty.vertex_array_buffers)
return;
gpu.dirty.vertex_array_buffers = false;
const auto& regs = gpu.regs; const auto& regs = gpu.regs;
MICROPROFILE_SCOPE(OpenGL_VB); MICROPROFILE_SCOPE(OpenGL_VB);
// Upload all guest vertex arrays sequentially to our buffer // Upload all guest vertex arrays sequentially to our buffer
for (u32 index = 0; index < Maxwell::NumVertexArrays; ++index) { for (u32 index = 0; index < Maxwell::NumVertexArrays; ++index) {
if (!gpu.dirty.vertex_array[index])
continue;
gpu.dirty.vertex_array[index] = false;
gpu.dirty.vertex_instance[index] = false;
const auto& vertex_array = regs.vertex_array[index]; const auto& vertex_array = regs.vertex_array[index];
if (!vertex_array.IsEnabled()) if (!vertex_array.IsEnabled())
continue; continue;
@ -224,19 +207,10 @@ void RasterizerOpenGL::SetupVertexBuffer(GLuint vao) {
void RasterizerOpenGL::SetupVertexInstances(GLuint vao) { void RasterizerOpenGL::SetupVertexInstances(GLuint vao) {
auto& gpu = system.GPU().Maxwell3D(); auto& gpu = system.GPU().Maxwell3D();
if (!gpu.dirty.vertex_instances)
return;
gpu.dirty.vertex_instances = false;
const auto& regs = gpu.regs; const auto& regs = gpu.regs;
// Upload all guest vertex arrays sequentially to our buffer // Upload all guest vertex arrays sequentially to our buffer
for (u32 index = 0; index < Maxwell::NumVertexArrays; ++index) { for (u32 index = 0; index < 16; ++index) {
if (!gpu.dirty.vertex_instance[index])
continue;
gpu.dirty.vertex_instance[index] = false;
if (regs.instanced_arrays.IsInstancingEnabled(index) && if (regs.instanced_arrays.IsInstancingEnabled(index) &&
regs.vertex_array[index].divisor != 0) { regs.vertex_array[index].divisor != 0) {
// Enable vertex buffer instancing with the specified divisor. // Enable vertex buffer instancing with the specified divisor.
@ -334,8 +308,6 @@ void RasterizerOpenGL::SetupShaders(GLenum primitive_mode) {
} }
SyncClipEnabled(clip_distances); SyncClipEnabled(clip_distances);
gpu.dirty.shaders = false;
} }
std::size_t RasterizerOpenGL::CalculateVertexArraysSize() const { std::size_t RasterizerOpenGL::CalculateVertexArraysSize() const {
@ -371,10 +343,6 @@ void RasterizerOpenGL::LoadDiskResources(const std::atomic_bool& stop_loading,
void RasterizerOpenGL::ConfigureFramebuffers() { void RasterizerOpenGL::ConfigureFramebuffers() {
MICROPROFILE_SCOPE(OpenGL_Framebuffer); MICROPROFILE_SCOPE(OpenGL_Framebuffer);
auto& gpu = system.GPU().Maxwell3D(); auto& gpu = system.GPU().Maxwell3D();
if (!gpu.dirty.render_settings) {
return;
}
gpu.dirty.render_settings = false;
texture_cache.GuardRenderTargets(true); texture_cache.GuardRenderTargets(true);
@ -453,7 +421,6 @@ void RasterizerOpenGL::Clear() {
OpenGLState prev_state{OpenGLState::GetCurState()}; OpenGLState prev_state{OpenGLState::GetCurState()};
SCOPE_EXIT({ SCOPE_EXIT({
prev_state.AllDirty();
prev_state.Apply(); prev_state.Apply();
}); });
@ -528,7 +495,6 @@ void RasterizerOpenGL::Clear() {
clear_state.EmulateViewportWithScissor(); clear_state.EmulateViewportWithScissor();
} }
clear_state.AllDirty();
clear_state.Apply(); clear_state.Apply();
if (use_color) { if (use_color) {
@ -631,12 +597,6 @@ void RasterizerOpenGL::Draw(bool is_indexed, bool is_instanced) {
bind_ubo_pushbuffer.Bind(); bind_ubo_pushbuffer.Bind();
bind_ssbo_pushbuffer.Bind(); bind_ssbo_pushbuffer.Bind();
if (invalidate) {
// As all cached buffers are invalidated, we need to recheck their state.
gpu.dirty.ResetVertexArrays();
}
gpu.dirty.memory_general = false;
shader_program_manager->ApplyTo(state); shader_program_manager->ApplyTo(state);
state.Apply(); state.Apply();
@ -1084,14 +1044,8 @@ void RasterizerOpenGL::SyncDepthTestState() {
void RasterizerOpenGL::SyncStencilTestState() { void RasterizerOpenGL::SyncStencilTestState() {
auto& maxwell3d = system.GPU().Maxwell3D(); auto& maxwell3d = system.GPU().Maxwell3D();
if (!maxwell3d.dirty.stencil_test) {
return;
}
maxwell3d.dirty.stencil_test = false;
const auto& regs = maxwell3d.regs; const auto& regs = maxwell3d.regs;
state.stencil.test_enabled = regs.stencil_enable != 0; state.stencil.test_enabled = regs.stencil_enable != 0;
state.MarkDirtyStencilState();
if (!regs.stencil_enable) { if (!regs.stencil_enable) {
return; return;
@ -1130,9 +1084,6 @@ void RasterizerOpenGL::SyncRasterizeEnable(OpenGLState& current_state) {
void RasterizerOpenGL::SyncColorMask() { void RasterizerOpenGL::SyncColorMask() {
auto& maxwell3d = system.GPU().Maxwell3D(); auto& maxwell3d = system.GPU().Maxwell3D();
if (!maxwell3d.dirty.color_mask) {
return;
}
const auto& regs = maxwell3d.regs; const auto& regs = maxwell3d.regs;
const std::size_t count = const std::size_t count =
@ -1145,9 +1096,6 @@ void RasterizerOpenGL::SyncColorMask() {
dest.blue_enabled = (source.B == 0) ? GL_FALSE : GL_TRUE; dest.blue_enabled = (source.B == 0) ? GL_FALSE : GL_TRUE;
dest.alpha_enabled = (source.A == 0) ? GL_FALSE : GL_TRUE; dest.alpha_enabled = (source.A == 0) ? GL_FALSE : GL_TRUE;
} }
state.MarkDirtyColorMask();
maxwell3d.dirty.color_mask = false;
} }
void RasterizerOpenGL::SyncMultiSampleState() { void RasterizerOpenGL::SyncMultiSampleState() {
@ -1163,9 +1111,6 @@ void RasterizerOpenGL::SyncFragmentColorClampState() {
void RasterizerOpenGL::SyncBlendState() { void RasterizerOpenGL::SyncBlendState() {
auto& maxwell3d = system.GPU().Maxwell3D(); auto& maxwell3d = system.GPU().Maxwell3D();
if (!maxwell3d.dirty.blend_state) {
return;
}
const auto& regs = maxwell3d.regs; const auto& regs = maxwell3d.regs;
state.blend_color.red = regs.blend_color.r; state.blend_color.red = regs.blend_color.r;
@ -1189,8 +1134,6 @@ void RasterizerOpenGL::SyncBlendState() {
for (std::size_t i = 1; i < Tegra::Engines::Maxwell3D::Regs::NumRenderTargets; i++) { for (std::size_t i = 1; i < Tegra::Engines::Maxwell3D::Regs::NumRenderTargets; i++) {
state.blend[i].enabled = false; state.blend[i].enabled = false;
} }
maxwell3d.dirty.blend_state = false;
state.MarkDirtyBlendState();
return; return;
} }
@ -1207,9 +1150,6 @@ void RasterizerOpenGL::SyncBlendState() {
blend.src_a_func = MaxwellToGL::BlendFunc(src.factor_source_a); blend.src_a_func = MaxwellToGL::BlendFunc(src.factor_source_a);
blend.dst_a_func = MaxwellToGL::BlendFunc(src.factor_dest_a); blend.dst_a_func = MaxwellToGL::BlendFunc(src.factor_dest_a);
} }
state.MarkDirtyBlendState();
maxwell3d.dirty.blend_state = false;
} }
void RasterizerOpenGL::SyncLogicOpState() { void RasterizerOpenGL::SyncLogicOpState() {
@ -1264,9 +1204,6 @@ void RasterizerOpenGL::SyncPointState() {
void RasterizerOpenGL::SyncPolygonOffset() { void RasterizerOpenGL::SyncPolygonOffset() {
auto& maxwell3d = system.GPU().Maxwell3D(); auto& maxwell3d = system.GPU().Maxwell3D();
if (!maxwell3d.dirty.polygon_offset) {
return;
}
const auto& regs = maxwell3d.regs; const auto& regs = maxwell3d.regs;
state.polygon_offset.fill_enable = regs.polygon_offset_fill_enable != 0; state.polygon_offset.fill_enable = regs.polygon_offset_fill_enable != 0;
@ -1277,9 +1214,6 @@ void RasterizerOpenGL::SyncPolygonOffset() {
state.polygon_offset.units = regs.polygon_offset_units / 2.0f; state.polygon_offset.units = regs.polygon_offset_units / 2.0f;
state.polygon_offset.factor = regs.polygon_offset_factor; state.polygon_offset.factor = regs.polygon_offset_factor;
state.polygon_offset.clamp = regs.polygon_offset_clamp; state.polygon_offset.clamp = regs.polygon_offset_clamp;
state.MarkDirtyPolygonOffset();
maxwell3d.dirty.polygon_offset = false;
} }
void RasterizerOpenGL::SyncAlphaTest() { void RasterizerOpenGL::SyncAlphaTest() {

View file

@ -623,10 +623,6 @@ bool ShaderCacheOpenGL::GenerateUnspecializedShaders(
} }
Shader ShaderCacheOpenGL::GetStageProgram(Maxwell::ShaderProgram program) { Shader ShaderCacheOpenGL::GetStageProgram(Maxwell::ShaderProgram program) {
if (!system.GPU().Maxwell3D().dirty.shaders) {
return last_shaders[static_cast<std::size_t>(program)];
}
auto& memory_manager{system.GPU().MemoryManager()}; auto& memory_manager{system.GPU().MemoryManager()};
const GPUVAddr address{GetShaderAddress(system, program)}; const GPUVAddr address{GetShaderAddress(system, program)};

View file

@ -189,11 +189,6 @@ void OpenGLState::ApplyRasterizerDiscard() {
} }
void OpenGLState::ApplyColorMask() { void OpenGLState::ApplyColorMask() {
if (!dirty.color_mask) {
return;
}
dirty.color_mask = false;
for (std::size_t i = 0; i < Maxwell::NumRenderTargets; ++i) { for (std::size_t i = 0; i < Maxwell::NumRenderTargets; ++i) {
const auto& updated = color_mask[i]; const auto& updated = color_mask[i];
auto& current = cur_state.color_mask[i]; auto& current = cur_state.color_mask[i];
@ -232,11 +227,6 @@ void OpenGLState::ApplyPrimitiveRestart() {
} }
void OpenGLState::ApplyStencilTest() { void OpenGLState::ApplyStencilTest() {
if (!dirty.stencil_state) {
return;
}
dirty.stencil_state = false;
Enable(GL_STENCIL_TEST, cur_state.stencil.test_enabled, stencil.test_enabled); Enable(GL_STENCIL_TEST, cur_state.stencil.test_enabled, stencil.test_enabled);
const auto ConfigStencil = [](GLenum face, const auto& config, auto& current) { const auto ConfigStencil = [](GLenum face, const auto& config, auto& current) {
@ -351,11 +341,6 @@ void OpenGLState::ApplyTargetBlending(std::size_t target, bool force) {
} }
void OpenGLState::ApplyBlending() { void OpenGLState::ApplyBlending() {
if (!dirty.blend_state) {
return;
}
dirty.blend_state = false;
if (independant_blend.enabled) { if (independant_blend.enabled) {
const bool force = independant_blend.enabled != cur_state.independant_blend.enabled; const bool force = independant_blend.enabled != cur_state.independant_blend.enabled;
for (std::size_t target = 0; target < Maxwell::NumRenderTargets; ++target) { for (std::size_t target = 0; target < Maxwell::NumRenderTargets; ++target) {
@ -383,11 +368,6 @@ void OpenGLState::ApplyLogicOp() {
} }
void OpenGLState::ApplyPolygonOffset() { void OpenGLState::ApplyPolygonOffset() {
if (!dirty.polygon_offset) {
return;
}
dirty.polygon_offset = false;
Enable(GL_POLYGON_OFFSET_FILL, cur_state.polygon_offset.fill_enable, Enable(GL_POLYGON_OFFSET_FILL, cur_state.polygon_offset.fill_enable,
polygon_offset.fill_enable); polygon_offset.fill_enable);
Enable(GL_POLYGON_OFFSET_LINE, cur_state.polygon_offset.line_enable, Enable(GL_POLYGON_OFFSET_LINE, cur_state.polygon_offset.line_enable,

View file

@ -212,39 +212,8 @@ public:
/// Viewport does not affects glClearBuffer so emulate viewport using scissor test /// Viewport does not affects glClearBuffer so emulate viewport using scissor test
void EmulateViewportWithScissor(); void EmulateViewportWithScissor();
void MarkDirtyBlendState() {
dirty.blend_state = true;
}
void MarkDirtyStencilState() {
dirty.stencil_state = true;
}
void MarkDirtyPolygonOffset() {
dirty.polygon_offset = true;
}
void MarkDirtyColorMask() {
dirty.color_mask = true;
}
void AllDirty() {
dirty.blend_state = true;
dirty.stencil_state = true;
dirty.polygon_offset = true;
dirty.color_mask = true;
}
private: private:
static OpenGLState cur_state; static OpenGLState cur_state;
struct {
bool blend_state;
bool stencil_state;
bool viewport_state;
bool polygon_offset;
bool color_mask;
} dirty{};
}; };
static_assert(std::is_trivially_copyable_v<OpenGLState>); static_assert(std::is_trivially_copyable_v<OpenGLState>);

View file

@ -522,7 +522,6 @@ void TextureCacheOpenGL::ImageBlit(View& src_view, View& dst_view,
OpenGLState prev_state{OpenGLState::GetCurState()}; OpenGLState prev_state{OpenGLState::GetCurState()};
SCOPE_EXIT({ SCOPE_EXIT({
prev_state.AllDirty();
prev_state.Apply(); prev_state.Apply();
}); });
@ -530,7 +529,6 @@ void TextureCacheOpenGL::ImageBlit(View& src_view, View& dst_view,
state.draw.read_framebuffer = src_framebuffer.handle; state.draw.read_framebuffer = src_framebuffer.handle;
state.draw.draw_framebuffer = dst_framebuffer.handle; state.draw.draw_framebuffer = dst_framebuffer.handle;
state.framebuffer_srgb.enabled = dst_params.srgb_conversion; state.framebuffer_srgb.enabled = dst_params.srgb_conversion;
state.AllDirty();
state.Apply(); state.Apply();
u32 buffers{}; u32 buffers{};

View file

@ -311,11 +311,6 @@ void RendererOpenGL::SwapBuffers(const Tegra::FramebufferConfig* framebuffer) {
return; return;
} }
// Maintain the rasterizer's state as a priority
OpenGLState prev_state = OpenGLState::GetCurState();
state.AllDirty();
state.Apply();
PrepareRendertarget(framebuffer); PrepareRendertarget(framebuffer);
RenderScreenshot(); RenderScreenshot();
@ -368,10 +363,6 @@ void RendererOpenGL::SwapBuffers(const Tegra::FramebufferConfig* framebuffer) {
m_current_frame++; m_current_frame++;
rasterizer->TickFrame(); rasterizer->TickFrame();
} }
// Restore the rasterizer state
prev_state.AllDirty();
prev_state.Apply();
} }
void RendererOpenGL::PrepareRendertarget(const Tegra::FramebufferConfig* framebuffer) { void RendererOpenGL::PrepareRendertarget(const Tegra::FramebufferConfig* framebuffer) {
@ -445,7 +436,6 @@ void RendererOpenGL::InitOpenGLObjects() {
// Link shaders and get variable locations // Link shaders and get variable locations
shader.CreateFromSource(vertex_shader, nullptr, fragment_shader); shader.CreateFromSource(vertex_shader, nullptr, fragment_shader);
state.draw.shader_program = shader.handle; state.draw.shader_program = shader.handle;
state.AllDirty();
state.Apply(); state.Apply();
// Generate VBO handle for drawing // Generate VBO handle for drawing
@ -580,14 +570,12 @@ void RendererOpenGL::DrawScreenTriangles(const ScreenInfo& screen_info, float x,
state.textures[0] = screen_info.display_texture; state.textures[0] = screen_info.display_texture;
state.framebuffer_srgb.enabled = screen_info.display_srgb; state.framebuffer_srgb.enabled = screen_info.display_srgb;
state.AllDirty();
state.Apply(); state.Apply();
glNamedBufferSubData(vertex_buffer.handle, 0, sizeof(vertices), std::data(vertices)); glNamedBufferSubData(vertex_buffer.handle, 0, sizeof(vertices), std::data(vertices));
glDrawArrays(GL_TRIANGLE_STRIP, 0, 4); glDrawArrays(GL_TRIANGLE_STRIP, 0, 4);
// Restore default state // Restore default state
state.framebuffer_srgb.enabled = false; state.framebuffer_srgb.enabled = false;
state.textures[0] = 0; state.textures[0] = 0;
state.AllDirty();
state.Apply(); state.Apply();
} }
@ -658,7 +646,6 @@ void RendererOpenGL::RenderScreenshot() {
GLuint old_read_fb = state.draw.read_framebuffer; GLuint old_read_fb = state.draw.read_framebuffer;
GLuint old_draw_fb = state.draw.draw_framebuffer; GLuint old_draw_fb = state.draw.draw_framebuffer;
state.draw.read_framebuffer = state.draw.draw_framebuffer = screenshot_framebuffer.handle; state.draw.read_framebuffer = state.draw.draw_framebuffer = screenshot_framebuffer.handle;
state.AllDirty();
state.Apply(); state.Apply();
Layout::FramebufferLayout layout{renderer_settings.screenshot_framebuffer_layout}; Layout::FramebufferLayout layout{renderer_settings.screenshot_framebuffer_layout};
@ -678,7 +665,6 @@ void RendererOpenGL::RenderScreenshot() {
screenshot_framebuffer.Release(); screenshot_framebuffer.Release();
state.draw.read_framebuffer = old_read_fb; state.draw.read_framebuffer = old_read_fb;
state.draw.draw_framebuffer = old_draw_fb; state.draw.draw_framebuffer = old_draw_fb;
state.AllDirty();
state.Apply(); state.Apply();
glDeleteRenderbuffers(1, &renderbuffer); glDeleteRenderbuffers(1, &renderbuffer);

View file

@ -172,11 +172,6 @@ VKPipelineCache::~VKPipelineCache() = default;
std::array<Shader, Maxwell::MaxShaderProgram> VKPipelineCache::GetShaders() { std::array<Shader, Maxwell::MaxShaderProgram> VKPipelineCache::GetShaders() {
const auto& gpu = system.GPU().Maxwell3D(); const auto& gpu = system.GPU().Maxwell3D();
auto& dirty = system.GPU().Maxwell3D().dirty.shaders;
if (!dirty) {
return last_shaders;
}
dirty = false;
std::array<Shader, Maxwell::MaxShaderProgram> shaders; std::array<Shader, Maxwell::MaxShaderProgram> shaders;
for (std::size_t index = 0; index < Maxwell::MaxShaderProgram; ++index) { for (std::size_t index = 0; index < Maxwell::MaxShaderProgram; ++index) {

View file

@ -568,9 +568,7 @@ void RasterizerVulkan::FlushWork() {
RasterizerVulkan::Texceptions RasterizerVulkan::UpdateAttachments() { RasterizerVulkan::Texceptions RasterizerVulkan::UpdateAttachments() {
MICROPROFILE_SCOPE(Vulkan_RenderTargets); MICROPROFILE_SCOPE(Vulkan_RenderTargets);
auto& dirty = system.GPU().Maxwell3D().dirty; constexpr bool update_rendertargets = true;
const bool update_rendertargets = dirty.render_settings;
dirty.render_settings = false;
texture_cache.GuardRenderTargets(true); texture_cache.GuardRenderTargets(true);
@ -973,10 +971,6 @@ void RasterizerVulkan::SetupImage(const Tegra::Texture::TICEntry& tic, const Ima
} }
void RasterizerVulkan::UpdateViewportsState(Tegra::Engines::Maxwell3D& gpu) { void RasterizerVulkan::UpdateViewportsState(Tegra::Engines::Maxwell3D& gpu) {
if (!gpu.dirty.viewport_transform && scheduler.TouchViewports()) {
return;
}
gpu.dirty.viewport_transform = false;
const auto& regs = gpu.regs; const auto& regs = gpu.regs;
const std::array viewports{ const std::array viewports{
GetViewportState(device, regs, 0), GetViewportState(device, regs, 1), GetViewportState(device, regs, 0), GetViewportState(device, regs, 1),
@ -993,10 +987,6 @@ void RasterizerVulkan::UpdateViewportsState(Tegra::Engines::Maxwell3D& gpu) {
} }
void RasterizerVulkan::UpdateScissorsState(Tegra::Engines::Maxwell3D& gpu) { void RasterizerVulkan::UpdateScissorsState(Tegra::Engines::Maxwell3D& gpu) {
if (!gpu.dirty.scissor_test && scheduler.TouchScissors()) {
return;
}
gpu.dirty.scissor_test = false;
const auto& regs = gpu.regs; const auto& regs = gpu.regs;
const std::array scissors = { const std::array scissors = {
GetScissorState(regs, 0), GetScissorState(regs, 1), GetScissorState(regs, 2), GetScissorState(regs, 0), GetScissorState(regs, 1), GetScissorState(regs, 2),
@ -1011,10 +1001,6 @@ void RasterizerVulkan::UpdateScissorsState(Tegra::Engines::Maxwell3D& gpu) {
} }
void RasterizerVulkan::UpdateDepthBias(Tegra::Engines::Maxwell3D& gpu) { void RasterizerVulkan::UpdateDepthBias(Tegra::Engines::Maxwell3D& gpu) {
if (!gpu.dirty.polygon_offset && scheduler.TouchDepthBias()) {
return;
}
gpu.dirty.polygon_offset = false;
const auto& regs = gpu.regs; const auto& regs = gpu.regs;
scheduler.Record([constant = regs.polygon_offset_units, clamp = regs.polygon_offset_clamp, scheduler.Record([constant = regs.polygon_offset_units, clamp = regs.polygon_offset_clamp,
factor = regs.polygon_offset_factor](auto cmdbuf, auto& dld) { factor = regs.polygon_offset_factor](auto cmdbuf, auto& dld) {
@ -1023,10 +1009,6 @@ void RasterizerVulkan::UpdateDepthBias(Tegra::Engines::Maxwell3D& gpu) {
} }
void RasterizerVulkan::UpdateBlendConstants(Tegra::Engines::Maxwell3D& gpu) { void RasterizerVulkan::UpdateBlendConstants(Tegra::Engines::Maxwell3D& gpu) {
if (!gpu.dirty.blend_state && scheduler.TouchBlendConstants()) {
return;
}
gpu.dirty.blend_state = false;
const std::array blend_color = {gpu.regs.blend_color.r, gpu.regs.blend_color.g, const std::array blend_color = {gpu.regs.blend_color.r, gpu.regs.blend_color.g,
gpu.regs.blend_color.b, gpu.regs.blend_color.a}; gpu.regs.blend_color.b, gpu.regs.blend_color.a};
scheduler.Record([blend_color](auto cmdbuf, auto& dld) { scheduler.Record([blend_color](auto cmdbuf, auto& dld) {
@ -1035,20 +1017,12 @@ void RasterizerVulkan::UpdateBlendConstants(Tegra::Engines::Maxwell3D& gpu) {
} }
void RasterizerVulkan::UpdateDepthBounds(Tegra::Engines::Maxwell3D& gpu) { void RasterizerVulkan::UpdateDepthBounds(Tegra::Engines::Maxwell3D& gpu) {
if (!gpu.dirty.depth_bounds_values && scheduler.TouchDepthBounds()) {
return;
}
gpu.dirty.depth_bounds_values = false;
const auto& regs = gpu.regs; const auto& regs = gpu.regs;
scheduler.Record([min = regs.depth_bounds[0], max = regs.depth_bounds[1]]( scheduler.Record([min = regs.depth_bounds[0], max = regs.depth_bounds[1]](
auto cmdbuf, auto& dld) { cmdbuf.setDepthBounds(min, max, dld); }); auto cmdbuf, auto& dld) { cmdbuf.setDepthBounds(min, max, dld); });
} }
void RasterizerVulkan::UpdateStencilFaces(Tegra::Engines::Maxwell3D& gpu) { void RasterizerVulkan::UpdateStencilFaces(Tegra::Engines::Maxwell3D& gpu) {
if (!gpu.dirty.stencil_test && scheduler.TouchStencilValues()) {
return;
}
gpu.dirty.stencil_test = false;
const auto& regs = gpu.regs; const auto& regs = gpu.regs;
if (regs.stencil_two_side_enable) { if (regs.stencil_two_side_enable) {
// Separate values per face // Separate values per face

View file

@ -143,11 +143,6 @@ public:
std::lock_guard lock{mutex}; std::lock_guard lock{mutex};
auto& maxwell3d = system.GPU().Maxwell3D(); auto& maxwell3d = system.GPU().Maxwell3D();
if (!maxwell3d.dirty.depth_buffer) {
return depth_buffer.view;
}
maxwell3d.dirty.depth_buffer = false;
const auto& regs{maxwell3d.regs}; const auto& regs{maxwell3d.regs};
const auto gpu_addr{regs.zeta.Address()}; const auto gpu_addr{regs.zeta.Address()};
if (!gpu_addr || !regs.zeta_enable) { if (!gpu_addr || !regs.zeta_enable) {
@ -175,10 +170,6 @@ public:
std::lock_guard lock{mutex}; std::lock_guard lock{mutex};
ASSERT(index < Tegra::Engines::Maxwell3D::Regs::NumRenderTargets); ASSERT(index < Tegra::Engines::Maxwell3D::Regs::NumRenderTargets);
auto& maxwell3d = system.GPU().Maxwell3D(); auto& maxwell3d = system.GPU().Maxwell3D();
if (!maxwell3d.dirty.render_target[index]) {
return render_targets[index].view;
}
maxwell3d.dirty.render_target[index] = false;
const auto& regs{maxwell3d.regs}; const auto& regs{maxwell3d.regs};
if (index >= regs.rt_control.count || regs.rt[index].Address() == 0 || if (index >= regs.rt_control.count || regs.rt[index].Address() == 0 ||
@ -319,16 +310,7 @@ protected:
// and reading it from a separate buffer. // and reading it from a separate buffer.
virtual void BufferCopy(TSurface& src_surface, TSurface& dst_surface) = 0; virtual void BufferCopy(TSurface& src_surface, TSurface& dst_surface) = 0;
void ManageRenderTargetUnregister(TSurface& surface) { void ManageRenderTargetUnregister([[maybe_unused]] TSurface& surface) {}
auto& maxwell3d = system.GPU().Maxwell3D();
const u32 index = surface->GetRenderTarget();
if (index == DEPTH_RT) {
maxwell3d.dirty.depth_buffer = true;
} else {
maxwell3d.dirty.render_target[index] = true;
}
maxwell3d.dirty.render_settings = true;
}
void Register(TSurface surface) { void Register(TSurface surface) {
const GPUVAddr gpu_addr = surface->GetGpuAddr(); const GPUVAddr gpu_addr = surface->GetGpuAddr();