Skip to content

Commit

Permalink
Backends: WebGPU: Use WGSL shaders instead of precompiled SPIR-V + ad…
Browse files Browse the repository at this point in the history 
…d Gamma uniform. (#6188)

Add gamma correction uniform
Group uniforms in a single binding
The second binding was not satisfying the minimum
BufferBindingType::Uniform alignment (256) and since this alignment is
large it is more idiomatic to group uniforms tegether.

Also ensures that the size of the uniform buffer is aligned to 16 bytes.
  • Loading branch information
@eliemichel @ocornut
eliemichel authored and ocornut committed Apr 11, 2023
1 parent e67f2f4 commit d81f2ae
Show file tree
Hide file tree
Showing 2 changed files with 92 additions and 115 deletions.
205 changes: 91 additions & 114 deletions backends/imgui_impl_wgpu.cpp
View file
Original file line number Diff line number Diff line change
Expand Up @@ -13,7 +13,7 @@

// CHANGELOG
// (minor and older changes stripped away, please see git history for details)
// 2023-04-11: Align buffer sizes.
// 2023-04-11: Align buffer sizes. Use WGSL shaders instead of precompiled SPIR-V.
// 2023-04-11: Reorganized backend to pull data from a single structure to facilitate usage with multiple-contexts (all g_XXXX access changed to bd->XXXX).
// 2023-01-25: Revert automatic pipeline layout generation (see https://github.com/gpuweb/gpuweb/issues/2470)
// 2022-11-24: Fixed validation error with default depth buffer settings.
Expand Down Expand Up @@ -62,6 +62,7 @@ struct FrameResources
struct Uniforms
{
float MVP[4][4];
float Gamma;
};

struct ImGui_ImplWGPU_Data
Expand Down Expand Up @@ -89,114 +90,60 @@ static ImGui_ImplWGPU_Data* ImGui_ImplWGPU_GetBackendData()
// SHADERS
//-----------------------------------------------------------------------------

// glsl_shader.vert, compiled with:
// # glslangValidator -V -x -o glsl_shader.vert.u32 glsl_shader.vert
/*
#version 450 core
layout(location = 0) in vec2 aPos;
layout(location = 1) in vec2 aUV;
layout(location = 2) in vec4 aColor;
layout(set=0, binding = 0) uniform transform { mat4 mvp; };
static const char __shader_vert_wgsl[] = R"(
struct VertexInput {
@location(0) position: vec2<f32>,
@location(1) uv: vec2<f32>,
@location(2) color: vec4<f32>,
};
out gl_PerVertex { vec4 gl_Position; };
layout(location = 0) out struct { vec4 Color; vec2 UV; } Out;
struct VertexOutput {
@builtin(position) position: vec4<f32>,
@location(0) color: vec4<f32>,
@location(1) uv: vec2<f32>,
};
void main()
{
Out.Color = aColor;
Out.UV = aUV;
gl_Position = mvp * vec4(aPos, 0, 1);
}
*/
static uint32_t __glsl_shader_vert_spv[] =
{
0x07230203,0x00010000,0x00080007,0x0000002c,0x00000000,0x00020011,0x00000001,0x0006000b,
0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001,
0x000a000f,0x00000000,0x00000004,0x6e69616d,0x00000000,0x0000000b,0x0000000f,0x00000015,
0x0000001b,0x00000023,0x00030003,0x00000002,0x000001c2,0x00040005,0x00000004,0x6e69616d,
0x00000000,0x00030005,0x00000009,0x00000000,0x00050006,0x00000009,0x00000000,0x6f6c6f43,
0x00000072,0x00040006,0x00000009,0x00000001,0x00005655,0x00030005,0x0000000b,0x0074754f,
0x00040005,0x0000000f,0x6c6f4361,0x0000726f,0x00030005,0x00000015,0x00565561,0x00060005,
0x00000019,0x505f6c67,0x65567265,0x78657472,0x00000000,0x00060006,0x00000019,0x00000000,
0x505f6c67,0x7469736f,0x006e6f69,0x00030005,0x0000001b,0x00000000,0x00050005,0x0000001d,
0x6e617274,0x726f6673,0x0000006d,0x00040006,0x0000001d,0x00000000,0x0070766d,0x00030005,
0x0000001f,0x00000000,0x00040005,0x00000023,0x736f5061,0x00000000,0x00040047,0x0000000b,
0x0000001e,0x00000000,0x00040047,0x0000000f,0x0000001e,0x00000002,0x00040047,0x00000015,
0x0000001e,0x00000001,0x00050048,0x00000019,0x00000000,0x0000000b,0x00000000,0x00030047,
0x00000019,0x00000002,0x00040048,0x0000001d,0x00000000,0x00000005,0x00050048,0x0000001d,
0x00000000,0x00000023,0x00000000,0x00050048,0x0000001d,0x00000000,0x00000007,0x00000010,
0x00030047,0x0000001d,0x00000002,0x00040047,0x0000001f,0x00000022,0x00000000,0x00040047,
0x0000001f,0x00000021,0x00000000,0x00040047,0x00000023,0x0000001e,0x00000000,0x00020013,
0x00000002,0x00030021,0x00000003,0x00000002,0x00030016,0x00000006,0x00000020,0x00040017,
0x00000007,0x00000006,0x00000004,0x00040017,0x00000008,0x00000006,0x00000002,0x0004001e,
0x00000009,0x00000007,0x00000008,0x00040020,0x0000000a,0x00000003,0x00000009,0x0004003b,
0x0000000a,0x0000000b,0x00000003,0x00040015,0x0000000c,0x00000020,0x00000001,0x0004002b,
0x0000000c,0x0000000d,0x00000000,0x00040020,0x0000000e,0x00000001,0x00000007,0x0004003b,
0x0000000e,0x0000000f,0x00000001,0x00040020,0x00000011,0x00000003,0x00000007,0x0004002b,
0x0000000c,0x00000013,0x00000001,0x00040020,0x00000014,0x00000001,0x00000008,0x0004003b,
0x00000014,0x00000015,0x00000001,0x00040020,0x00000017,0x00000003,0x00000008,0x0003001e,
0x00000019,0x00000007,0x00040020,0x0000001a,0x00000003,0x00000019,0x0004003b,0x0000001a,
0x0000001b,0x00000003,0x00040018,0x0000001c,0x00000007,0x00000004,0x0003001e,0x0000001d,
0x0000001c,0x00040020,0x0000001e,0x00000002,0x0000001d,0x0004003b,0x0000001e,0x0000001f,
0x00000002,0x00040020,0x00000020,0x00000002,0x0000001c,0x0004003b,0x00000014,0x00000023,
0x00000001,0x0004002b,0x00000006,0x00000025,0x00000000,0x0004002b,0x00000006,0x00000026,
0x3f800000,0x00050036,0x00000002,0x00000004,0x00000000,0x00000003,0x000200f8,0x00000005,
0x0004003d,0x00000007,0x00000010,0x0000000f,0x00050041,0x00000011,0x00000012,0x0000000b,
0x0000000d,0x0003003e,0x00000012,0x00000010,0x0004003d,0x00000008,0x00000016,0x00000015,
0x00050041,0x00000017,0x00000018,0x0000000b,0x00000013,0x0003003e,0x00000018,0x00000016,
0x00050041,0x00000020,0x00000021,0x0000001f,0x0000000d,0x0004003d,0x0000001c,0x00000022,
0x00000021,0x0004003d,0x00000008,0x00000024,0x00000023,0x00050051,0x00000006,0x00000027,
0x00000024,0x00000000,0x00050051,0x00000006,0x00000028,0x00000024,0x00000001,0x00070050,
0x00000007,0x00000029,0x00000027,0x00000028,0x00000025,0x00000026,0x00050091,0x00000007,
0x0000002a,0x00000022,0x00000029,0x00050041,0x00000011,0x0000002b,0x0000001b,0x0000000d,
0x0003003e,0x0000002b,0x0000002a,0x000100fd,0x00010038
struct Uniforms {
mvp: mat4x4<f32>,
gamma: f32,
};
// glsl_shader.frag, compiled with:
// # glslangValidator -V -x -o glsl_shader.frag.u32 glsl_shader.frag
/*
#version 450 core
layout(location = 0) out vec4 fColor;
layout(set=0, binding=1) uniform sampler s;
layout(set=1, binding=0) uniform texture2D t;
layout(location = 0) in struct { vec4 Color; vec2 UV; } In;
void main()
{
fColor = In.Color * texture(sampler2D(t, s), In.UV.st);
@group(0) @binding(0) var<uniform> uniforms: Uniforms;
@vertex
fn main(in: VertexInput) -> VertexOutput {
var out: VertexOutput;
out.position = uniforms.mvp * vec4<f32>(in.position, 0.0, 1.0);
out.color = in.color;
out.uv = in.uv;
return out;
}
*/
static uint32_t __glsl_shader_frag_spv[] =
{
0x07230203,0x00010000,0x00080007,0x00000023,0x00000000,0x00020011,0x00000001,0x0006000b,
0x00000001,0x4c534c47,0x6474732e,0x3035342e,0x00000000,0x0003000e,0x00000000,0x00000001,
0x0007000f,0x00000004,0x00000004,0x6e69616d,0x00000000,0x00000009,0x0000000d,0x00030010,
0x00000004,0x00000007,0x00030003,0x00000002,0x000001c2,0x00040005,0x00000004,0x6e69616d,
0x00000000,0x00040005,0x00000009,0x6c6f4366,0x0000726f,0x00030005,0x0000000b,0x00000000,
0x00050006,0x0000000b,0x00000000,0x6f6c6f43,0x00000072,0x00040006,0x0000000b,0x00000001,
0x00005655,0x00030005,0x0000000d,0x00006e49,0x00030005,0x00000015,0x00000074,0x00030005,
0x00000019,0x00000073,0x00040047,0x00000009,0x0000001e,0x00000000,0x00040047,0x0000000d,
0x0000001e,0x00000000,0x00040047,0x00000015,0x00000022,0x00000001,0x00040047,0x00000015,
0x00000021,0x00000000,0x00040047,0x00000019,0x00000022,0x00000000,0x00040047,0x00000019,
0x00000021,0x00000001,0x00020013,0x00000002,0x00030021,0x00000003,0x00000002,0x00030016,
0x00000006,0x00000020,0x00040017,0x00000007,0x00000006,0x00000004,0x00040020,0x00000008,
0x00000003,0x00000007,0x0004003b,0x00000008,0x00000009,0x00000003,0x00040017,0x0000000a,
0x00000006,0x00000002,0x0004001e,0x0000000b,0x00000007,0x0000000a,0x00040020,0x0000000c,
0x00000001,0x0000000b,0x0004003b,0x0000000c,0x0000000d,0x00000001,0x00040015,0x0000000e,
0x00000020,0x00000001,0x0004002b,0x0000000e,0x0000000f,0x00000000,0x00040020,0x00000010,
0x00000001,0x00000007,0x00090019,0x00000013,0x00000006,0x00000001,0x00000000,0x00000000,
0x00000000,0x00000001,0x00000000,0x00040020,0x00000014,0x00000000,0x00000013,0x0004003b,
0x00000014,0x00000015,0x00000000,0x0002001a,0x00000017,0x00040020,0x00000018,0x00000000,
0x00000017,0x0004003b,0x00000018,0x00000019,0x00000000,0x0003001b,0x0000001b,0x00000013,
0x0004002b,0x0000000e,0x0000001d,0x00000001,0x00040020,0x0000001e,0x00000001,0x0000000a,
0x00050036,0x00000002,0x00000004,0x00000000,0x00000003,0x000200f8,0x00000005,0x00050041,
0x00000010,0x00000011,0x0000000d,0x0000000f,0x0004003d,0x00000007,0x00000012,0x00000011,
0x0004003d,0x00000013,0x00000016,0x00000015,0x0004003d,0x00000017,0x0000001a,0x00000019,
0x00050056,0x0000001b,0x0000001c,0x00000016,0x0000001a,0x00050041,0x0000001e,0x0000001f,
0x0000000d,0x0000001d,0x0004003d,0x0000000a,0x00000020,0x0000001f,0x00050057,0x00000007,
0x00000021,0x0000001c,0x00000020,0x00050085,0x00000007,0x00000022,0x00000012,0x00000021,
0x0003003e,0x00000009,0x00000022,0x000100fd,0x00010038
)";

static const char __shader_frag_wgsl[] = R"(
struct VertexOutput {
@builtin(position) position: vec4<f32>,
@location(0) color: vec4<f32>,
@location(1) uv: vec2<f32>,
};
struct Uniforms {
mvp: mat4x4<f32>,
gamma: f32,
};
@group(0) @binding(0) var<uniform> uniforms: Uniforms;
@group(0) @binding(1) var s: sampler;
@group(1) @binding(0) var t: texture_2d<f32>;
@fragment
fn main(in: VertexOutput) -> @location(0) vec4<f32> {
let color = in.color * textureSample(t, s, in.uv);
let corrected_color = pow(color.rgb, vec3<f32>(uniforms.gamma));
return vec4<f32>(corrected_color, color.a);
}
)";

static void SafeRelease(ImDrawIdx*& res)
{
if (res)
Expand Down Expand Up @@ -277,17 +224,16 @@ static void SafeRelease(FrameResources& res)
SafeRelease(res.VertexBufferHost);
}

static WGPUProgrammableStageDescriptor ImGui_ImplWGPU_CreateShaderModule(uint32_t* binary_data, uint32_t binary_data_size)
static WGPUProgrammableStageDescriptor ImGui_ImplWGPU_CreateShaderModule(const char* wgsl_source)
{
ImGui_ImplWGPU_Data* bd = ImGui_ImplWGPU_GetBackendData();

WGPUShaderModuleSPIRVDescriptor spirv_desc = {};
spirv_desc.chain.sType = WGPUSType_ShaderModuleSPIRVDescriptor;
spirv_desc.codeSize = binary_data_size;
spirv_desc.code = binary_data;
WGPUShaderModuleWGSLDescriptor wgsl_desc = {};
wgsl_desc.chain.sType = WGPUSType_ShaderModuleWGSLDescriptor;
wgsl_desc.source = wgsl_source;

WGPUShaderModuleDescriptor desc = {};
desc.nextInChain = reinterpret_cast<WGPUChainedStruct*>(&spirv_desc);
desc.nextInChain = reinterpret_cast<WGPUChainedStruct*>(&wgsl_desc);

WGPUProgrammableStageDescriptor stage_desc = {};
stage_desc.module = wgpuDeviceCreateShaderModule(bd->wgpuDevice, &desc);
Expand Down Expand Up @@ -325,7 +271,38 @@ static void ImGui_ImplWGPU_SetupRenderState(ImDrawData* draw_data, WGPURenderPas
{ 0.0f, 0.0f, 0.5f, 0.0f },
{ (R+L)/(L-R), (T+B)/(B-T), 0.5f, 1.0f },
};
wgpuQueueWriteBuffer(bd->defaultQueue, bd->renderResources.Uniforms, 0, mvp, sizeof(mvp));
wgpuQueueWriteBuffer(bd->defaultQueue, bd->renderResources.Uniforms, offsetof(Uniforms, MVP), mvp, sizeof(Uniforms::MVP));
float gamma;
switch (bd->renderTargetFormat)
{
case WGPUTextureFormat_ASTC10x10UnormSrgb:
case WGPUTextureFormat_ASTC10x5UnormSrgb:
case WGPUTextureFormat_ASTC10x6UnormSrgb:
case WGPUTextureFormat_ASTC10x8UnormSrgb:
case WGPUTextureFormat_ASTC12x10UnormSrgb:
case WGPUTextureFormat_ASTC12x12UnormSrgb:
case WGPUTextureFormat_ASTC4x4UnormSrgb:
case WGPUTextureFormat_ASTC5x5UnormSrgb:
case WGPUTextureFormat_ASTC6x5UnormSrgb:
case WGPUTextureFormat_ASTC6x6UnormSrgb:
case WGPUTextureFormat_ASTC8x5UnormSrgb:
case WGPUTextureFormat_ASTC8x6UnormSrgb:
case WGPUTextureFormat_ASTC8x8UnormSrgb:
case WGPUTextureFormat_BC1RGBAUnormSrgb:
case WGPUTextureFormat_BC2RGBAUnormSrgb:
case WGPUTextureFormat_BC3RGBAUnormSrgb:
case WGPUTextureFormat_BC7RGBAUnormSrgb:
case WGPUTextureFormat_BGRA8UnormSrgb:
case WGPUTextureFormat_ETC2RGB8A1UnormSrgb:
case WGPUTextureFormat_ETC2RGB8UnormSrgb:
case WGPUTextureFormat_ETC2RGBA8UnormSrgb:
case WGPUTextureFormat_RGBA8UnormSrgb:
gamma = 2.2f;
break;
default:
gamma = 1.0f;
}
wgpuQueueWriteBuffer(bd->defaultQueue, bd->renderResources.Uniforms, offsetof(Uniforms, Gamma), &gamma, sizeof(Uniforms::Gamma));
}

// Setup viewport
Expand Down Expand Up @@ -555,7 +532,7 @@ static void ImGui_ImplWGPU_CreateUniformBuffer()
nullptr,
"Dear ImGui Uniform buffer",
WGPUBufferUsage_CopyDst | WGPUBufferUsage_Uniform,
MEMALIGN(sizeof(Uniforms), 4),
MEMALIGN(sizeof(Uniforms), 16),
false
};
bd->renderResources.Uniforms = wgpuDeviceCreateBuffer(bd->wgpuDevice, &ub_desc);
Expand All @@ -582,7 +559,7 @@ bool ImGui_ImplWGPU_CreateDeviceObjects()
// Bind group layouts
WGPUBindGroupLayoutEntry common_bg_layout_entries[2] = {};
common_bg_layout_entries[0].binding = 0;
common_bg_layout_entries[0].visibility = WGPUShaderStage_Vertex;
common_bg_layout_entries[0].visibility = WGPUShaderStage_Vertex | WGPUShaderStage_Fragment;
common_bg_layout_entries[0].buffer.type = WGPUBufferBindingType_Uniform;
common_bg_layout_entries[1].binding = 1;
common_bg_layout_entries[1].visibility = WGPUShaderStage_Fragment;
Expand Down Expand Up @@ -612,7 +589,7 @@ bool ImGui_ImplWGPU_CreateDeviceObjects()
graphics_pipeline_desc.layout = wgpuDeviceCreatePipelineLayout(bd->wgpuDevice, &layout_desc);

// Create the vertex shader
WGPUProgrammableStageDescriptor vertex_shader_desc = ImGui_ImplWGPU_CreateShaderModule(__glsl_shader_vert_spv, sizeof(__glsl_shader_vert_spv) / sizeof(uint32_t));
WGPUProgrammableStageDescriptor vertex_shader_desc = ImGui_ImplWGPU_CreateShaderModule(__shader_vert_wgsl);
graphics_pipeline_desc.vertex.module = vertex_shader_desc.module;
graphics_pipeline_desc.vertex.entryPoint = vertex_shader_desc.entryPoint;

Expand All @@ -634,7 +611,7 @@ bool ImGui_ImplWGPU_CreateDeviceObjects()
graphics_pipeline_desc.vertex.buffers = buffer_layouts;

// Create the pixel shader
WGPUProgrammableStageDescriptor pixel_shader_desc = ImGui_ImplWGPU_CreateShaderModule(__glsl_shader_frag_spv, sizeof(__glsl_shader_frag_spv) / sizeof(uint32_t));
WGPUProgrammableStageDescriptor pixel_shader_desc = ImGui_ImplWGPU_CreateShaderModule(__shader_frag_wgsl);

// Create the blending setup
WGPUBlendState blend_state = {};
Expand Down Expand Up @@ -677,7 +654,7 @@ bool ImGui_ImplWGPU_CreateDeviceObjects()
// Create resource bind group
WGPUBindGroupEntry common_bg_entries[] =
{
{ nullptr, 0, bd->renderResources.Uniforms, 0, sizeof(Uniforms), 0, 0 },
{ nullptr, 0, bd->renderResources.Uniforms, 0, MEMALIGN(sizeof(Uniforms), 16), 0, 0 },
{ nullptr, 1, 0, 0, 0, bd->renderResources.Sampler, 0 },
};

Expand Down
2 changes: 1 addition & 1 deletion docs/CHANGELOG.txt
View file
Original file line number Diff line number Diff line change
Expand Up @@ -90,7 +90,7 @@ Other changes:
(#6315) [@PathogenDavid]
- Backends: OSX: Added support for io.AddMouseSourceEvent() to discriminate Mouse/Pen.
(#6314) [@PathogenDavid]
- Backends: WebGPU: Align buffers. (#6188) [@eliemichel]
- Backends: WebGPU: Align buffers. Use WGSL shaders instead of SPIR-V. Add gamma uniform. (#6188) [@eliemichel]
- Backends: WebGPU: Reorganized to store data in io.BackendRendererUserData like other backends.
- Examples: Windows: Added 'misc/debuggers/imgui.natstepfilter' file to all Visual Studio projects,
now that VS 2022 17.6 Preview 2 support adding Debug Step Filter spec files into projects.
Expand Down

0 comments on commit d81f2ae

Please sign in to comment.