starred/duckstation
1
0
Fork 0
mirror of https://github.com/stenzek/duckstation.git synced 2026-05-21 07:26:42 +00:00
Code Issues 1.1k Packages Projects Releases 20 Wiki Activity

PostProcessing/Slang: Separate original and input textures

Browse Source 
Only applies when chaining.
This commit is contained in:
Stenzek
2025-11-23 12:38:18 +10:00
parent 5d9309cd55
commit c9c4997d6f
8 changed files with 58 additions and 49 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
src/util/postprocessing.cpp
View File

@@ -826,6 +826,7 @@ GPUDevice::PresentResult PostProcessing::Chain::Apply(GPUTexture* input_color, G
{
GL_SCOPE_FMT("{} Apply", m_section);
GPUTexture* original_color = input_color;
GPUTexture* output = m_output_texture.get();
input_color->MakeReadyForSampling();
if (input_depth)
@@ -837,9 +838,9 @@ GPUDevice::PresentResult PostProcessing::Chain::Apply(GPUTexture* input_color, G
if (!stage->IsEnabled())
continue;
if (const GPUDevice::PresentResult pres =
stage->Apply(input_color, input_depth, stage->IsFinalStage() ? final_target : output, final_rect, orig_width,
orig_height, native_width, native_height, m_target_width, m_target_height, time);
if (const GPUDevice::PresentResult pres = stage->Apply(
original_color, input_color, input_depth, stage->IsFinalStage() ? final_target : output, final_rect,
orig_width, orig_height, native_width, native_height, m_target_width, m_target_height, time);
pres != GPUDevice::PresentResult::OK)
{
return pres;

7
src/util/postprocessing_shader.h
View File

@@ -54,9 +54,10 @@ public:
virtual bool CompilePipeline(GPUTexture::Format format, u32 width, u32 height, Error* error,
ProgressCallback* progress) = 0;
virtual GPUDevice::PresentResult Apply(GPUTexture* input_color, GPUTexture* input_depth, GPUTexture* final_target,
GSVector4i final_rect, s32 orig_width, s32 orig_height, s32 native_width,
s32 native_height, u32 target_width, u32 target_height, float time) = 0;
virtual GPUDevice::PresentResult Apply(GPUTexture* original_color, GPUTexture* input_color, GPUTexture* input_depth,
GPUTexture* final_target, GSVector4i final_rect, s32 orig_width,
s32 orig_height, s32 native_width, s32 native_height, u32 target_width,
u32 target_height, float time) = 0;
protected:
using OptionList = std::vector<ShaderOption>;

10
src/util/postprocessing_shader_fx.cpp
View File

@@ -1602,11 +1602,11 @@ bool PostProcessing::ReShadeFXShader::ResizeTargets(u32 source_width, u32 source
return true;
}
GPUDevice::PresentResult PostProcessing::ReShadeFXShader::Apply(GPUTexture* input_color, GPUTexture* input_depth,
GPUTexture* final_target, GSVector4i final_rect,
s32 orig_width, s32 orig_height, s32 native_width,
s32 native_height, u32 target_width, u32 target_height,
float time)
GPUDevice::PresentResult PostProcessing::ReShadeFXShader::Apply(GPUTexture* original_color, GPUTexture* input_color,
GPUTexture* input_depth, GPUTexture* final_target,
GSVector4i final_rect, s32 orig_width, s32 orig_height,
s32 native_width, s32 native_height, u32 target_width,
u32 target_height, float time)
{
GL_SCOPE_FMT("PostProcessingShaderFX {}", m_name);

7
src/util/postprocessing_shader_fx.h
View File

@@ -36,9 +36,10 @@ public:
bool CompilePipeline(GPUTexture::Format format, u32 width, u32 height, Error* error,
ProgressCallback* progress) override;
GPUDevice::PresentResult Apply(GPUTexture* input_color, GPUTexture* input_depth, GPUTexture* final_target,
GSVector4i final_rect, s32 orig_width, s32 orig_height, s32 native_width,
s32 native_height, u32 target_width, u32 target_height, float time) override;
GPUDevice::PresentResult Apply(GPUTexture* original_color, GPUTexture* input_color, GPUTexture* input_depth,
GPUTexture* final_target, GSVector4i final_rect, s32 orig_width, s32 orig_height,
s32 native_width, s32 native_height, u32 target_width, u32 target_height,
float time) override;
private:
using TextureID = s32;

10
src/util/postprocessing_shader_glsl.cpp
View File

@@ -161,11 +161,11 @@ bool PostProcessing::GLSLShader::CompilePipeline(GPUTexture::Format format, u32
return true;
}
GPUDevice::PresentResult PostProcessing::GLSLShader::Apply(GPUTexture* input_color, GPUTexture* input_depth,
GPUTexture* final_target, GSVector4i final_rect,
s32 orig_width, s32 orig_height, s32 native_width,
s32 native_height, u32 target_width, u32 target_height,
float time)
GPUDevice::PresentResult PostProcessing::GLSLShader::Apply(GPUTexture* original_color, GPUTexture* input_color,
GPUTexture* input_depth, GPUTexture* final_target,
GSVector4i final_rect, s32 orig_width, s32 orig_height,
s32 native_width, s32 native_height, u32 target_width,
u32 target_height, float time)
{
GL_SCOPE_FMT("GLSL Shader {}", m_name);

7
src/util/postprocessing_shader_glsl.h
View File

@@ -22,9 +22,10 @@ public:
bool CompilePipeline(GPUTexture::Format format, u32 width, u32 height, Error* error,
ProgressCallback* progress) override;
GPUDevice::PresentResult Apply(GPUTexture* input_color, GPUTexture* input_depth, GPUTexture* final_target,
GSVector4i final_rect, s32 orig_width, s32 orig_height, s32 native_width,
s32 native_height, u32 target_width, u32 target_height, float time) override;
GPUDevice::PresentResult Apply(GPUTexture* original_color, GPUTexture* input_color, GPUTexture* input_depth,
GPUTexture* final_target, GSVector4i final_rect, s32 orig_width, s32 orig_height,
s32 native_width, s32 native_height, u32 target_width, u32 target_height,
float time) override;
private:
struct CommonUniforms

42
src/util/postprocessing_shader_slang.cpp
View File

@@ -1755,11 +1755,11 @@ std::unique_ptr<GPUPipeline> PostProcessing::SlangShader::CreateBlitPipeline(GPU
return g_gpu_device->CreatePipeline(plconfig, error);
}
GPUDevice::PresentResult PostProcessing::SlangShader::Apply(GPUTexture* input_color, GPUTexture* input_depth,
GPUTexture* final_target, GSVector4i final_rect,
s32 orig_width, s32 orig_height, s32 native_width,
s32 native_height, u32 target_width, u32 target_height,
float time)
GPUDevice::PresentResult PostProcessing::SlangShader::Apply(GPUTexture* original_color, GPUTexture* input_color,
GPUTexture* input_depth, GPUTexture* final_target,
GSVector4i final_rect, s32 orig_width, s32 orig_height,
s32 native_width, s32 native_height, u32 target_width,
u32 target_height, float time)
{
const auto bind_final_target = [](GPUTexture* final_target, const GSVector4i& final_rect) {
if (!final_target)
@@ -1793,16 +1793,16 @@ GPUDevice::PresentResult PostProcessing::SlangShader::Apply(GPUTexture* input_co
// Need to copy the input before drawing, because of swap chain
Error error;
if (!g_gpu_device->ResizeTexture(&m_original_history_textures[0], input_color->GetWidth(), input_color->GetHeight(),
GPUTexture::Type::Texture, input_color->GetFormat(), GPUTexture::Flags::None,
false, &error))
if (!g_gpu_device->ResizeTexture(&m_original_history_textures[0], original_color->GetWidth(),
original_color->GetHeight(), GPUTexture::Type::Texture,
original_color->GetFormat(), GPUTexture::Flags::None, false, &error))
{
ERROR_LOG("Failed to resize original history texture: {}", error.GetDescription());
}
else
{
g_gpu_device->CopyTextureRegion(m_original_history_textures[0].get(), 0, 0, 0, 0, input_color, 0, 0, 0, 0,
input_color->GetWidth(), input_color->GetHeight());
g_gpu_device->CopyTextureRegion(m_original_history_textures[0].get(), 0, 0, 0, 0, original_color, 0, 0, 0, 0,
original_color->GetWidth(), original_color->GetHeight());
}
}
@@ -1852,7 +1852,7 @@ GPUDevice::PresentResult PostProcessing::SlangShader::Apply(GPUTexture* input_co
for (const auto& [tex_id, bind_point] : pass.samplers)
{
GL_INS_FMT("Texture {}: ID {} [{}]", bind_point, tex_id, GetTextureNameForID(tex_id));
std::tie(textures[bind_point], samplers[bind_point]) = GetTextureByID(pass, tex_id, input_color);
std::tie(textures[bind_point], samplers[bind_point]) = GetTextureByID(pass, tex_id, original_color, input_color);
if (textures[bind_point])
textures[bind_point]->MakeReadyForSampling();
}
@@ -1891,8 +1891,8 @@ GPUDevice::PresentResult PostProcessing::SlangShader::Apply(GPUTexture* input_co
g_gpu_device->SetTextureSampler(static_cast<u32>(j), textures[j], samplers[j]);
alignas(VECTOR_ALIGNMENT) u8 push_constant_data[MAX_PUSH_CONSTANT_SIZE];
BindPassUniforms(pass, push_constant_data, input_color, output_size, final_rect, orig_width, orig_height,
native_width, native_height, target_width, target_height, time);
BindPassUniforms(pass, push_constant_data, original_color, input_color, output_size, final_rect, orig_width,
orig_height, native_width, native_height, target_width, target_height, time);
if (pass.push_constants_size > 0)
g_gpu_device->DrawWithPushConstants(4, base_vertex, push_constant_data, pass.push_constants_size);
else
@@ -2041,6 +2041,7 @@ TinyString PostProcessing::SlangShader::GetTextureNameForID(TextureID id) const
}
std::tuple<GPUTexture*, GPUSampler*> PostProcessing::SlangShader::GetTextureByID(const Pass& pass, TextureID id,
GPUTexture* original_color,
GPUTexture* input_color) const
{
if (id == TEXTURE_ID_SOURCE)
@@ -2054,12 +2055,12 @@ std::tuple<GPUTexture*, GPUSampler*> PostProcessing::SlangShader::GetTextureByID
else
{
// First pass, no last texture.
id = TEXTURE_ID_ORIGINAL;
return {input_color, g_gpu_device->GetNearestSampler()};
}
}
if (id == TEXTURE_ID_ORIGINAL)
return {input_color, g_gpu_device->GetNearestSampler()};
return {original_color, g_gpu_device->GetNearestSampler()};
if (id >= TEXTURE_ID_ORIGINAL_HISTORY_START)
{
@@ -2078,10 +2079,10 @@ std::tuple<GPUTexture*, GPUSampler*> PostProcessing::SlangShader::GetTextureByID
}
void PostProcessing::SlangShader::BindPassUniforms(const Pass& pass, u8* const push_constant_data,
GPUTexture* input_color, GSVector2i output_size,
GSVector4i final_rect, s32 orig_width, s32 orig_height,
s32 native_width, s32 native_height, u32 target_width,
u32 target_height, float time)
GPUTexture* original_color, GPUTexture* input_color,
GSVector2i output_size, GSVector4i final_rect, s32 orig_width,
s32 orig_height, s32 native_width, s32 native_height,
u32 target_width, u32 target_height, float time)
{
GL_INS_FMT("Uniform buffer: {} bytes", pass.uniforms_size);
GL_INS_FMT("Push constants: {} bytes", pass.push_constants_size);
@@ -2132,7 +2133,8 @@ void PostProcessing::SlangShader::BindPassUniforms(const Pass& pass, u8* const p
case BuiltinUniform::TextureSize:
{
const GPUTexture* texture = std::get<0>(GetTextureByID(pass, ui.associated_texture, input_color));
const GPUTexture* texture =
std::get<0>(GetTextureByID(pass, ui.associated_texture, original_color, input_color));
DebugAssert(texture);
const GSVector2 v = GSVector2(texture->GetSizeVec());

17
src/util/postprocessing_shader_slang.h
View File

@@ -34,9 +34,10 @@ public:
u32 target_height, u32 viewport_width, u32 viewport_height, Error* error) override;
bool CompilePipeline(GPUTexture::Format format, u32 width, u32 height, Error* error,
ProgressCallback* progress) override;
GPUDevice::PresentResult Apply(GPUTexture* input_color, GPUTexture* input_depth, GPUTexture* final_target,
GSVector4i final_rect, s32 orig_width, s32 orig_height, s32 native_width,
s32 native_height, u32 target_width, u32 target_height, float time) override;
GPUDevice::PresentResult Apply(GPUTexture* original_color, GPUTexture* input_color, GPUTexture* input_depth,
GPUTexture* final_target, GSVector4i final_rect, s32 orig_width, s32 orig_height,
s32 native_width, s32 native_height, u32 target_width, u32 target_height,
float time) override;
private:
using TextureID = u32;
@@ -129,11 +130,13 @@ private:
std::optional<Uniform> GetUniformInfo(std::string_view name, bool push_constant, u32 offset, u16 size, Error* error);
TinyString GetTextureNameForID(TextureID id) const;
std::tuple<GPUTexture*, GPUSampler*> GetTextureByID(const Pass& pass, TextureID id, GPUTexture* input_color) const;
std::tuple<GPUTexture*, GPUSampler*> GetTextureByID(const Pass& pass, TextureID id, GPUTexture* original_color,
GPUTexture* input_color) const;
void BindPassUniforms(const Pass& pass, u8* const push_constant_data, GPUTexture* input_color, GSVector2i output_size,
GSVector4i final_rect, s32 orig_width, s32 orig_height, s32 native_width, s32 native_height,
u32 target_width, u32 target_height, float time);
void BindPassUniforms(const Pass& pass, u8* const push_constant_data, GPUTexture* original_color,
GPUTexture* input_color, GSVector2i output_size, GSVector4i final_rect, s32 orig_width,
s32 orig_height, s32 native_width, s32 native_height, u32 target_width, u32 target_height,
float time);
std::vector<Texture> m_textures;
std::vector<Pass> m_passes;