mirror of
https://github.com/encounter/dawn-cmake.git
synced 2025-07-13 08:36:08 +00:00
a5d73ce965
This is a major reworking of this transform. The old transform code was getting unwieldy, with part of the complication coming from the handling of multiple return statements. By generating a wrapper function instead, we can avoid a lot of this complexity. The original entry point function is stripped of all shader IO attributes (as well as `stage` and `workgroup_size`), but the body is left unmodified. A new entry point wrapper function is introduced which calls the original function, packing/unpacking the shader inputs as necessary, and propagates the result to the corresponding shader outputs. The new code has been refactored to use a state object with the different parts of the transform split into separate functions, which makes it much more manageable. Fixed: tint:1076 Bug: tint:920 Change-Id: I3490a0ea7a3509a4e198ce730e476516649d8d96 Reviewed-on: https://dawn-review.googlesource.com/c/tint/+/60521 Auto-Submit: James Price <jrprice@google.com> Kokoro: Kokoro <noreply+kokoro@google.com> Commit-Queue: James Price <jrprice@google.com> Reviewed-by: Ben Clayton <bclayton@google.com>
47 lines
1.8 KiB
Plaintext
47 lines
1.8 KiB
Plaintext
#include <metal_stdlib>
|
|
|
|
using namespace metal;
|
|
struct Uniforms {
|
|
/* 0x0000 */ uint dstTextureFlipY;
|
|
/* 0x0004 */ uint isFloat16;
|
|
/* 0x0008 */ uint isRGB10A2Unorm;
|
|
/* 0x000c */ uint channelCount;
|
|
};
|
|
struct OutputBuf {
|
|
/* 0x0000 */ uint result[1];
|
|
};
|
|
|
|
uint ConvertToFp16FloatValue(float fp32) {
|
|
return 1u;
|
|
}
|
|
|
|
void tint_symbol_inner(constant Uniforms& uniforms, device OutputBuf& output, uint3 GlobalInvocationID, texture2d<float, access::sample> tint_symbol_1, texture2d<float, access::sample> tint_symbol_2) {
|
|
int2 size = int2(tint_symbol_1.get_width(), tint_symbol_1.get_height());
|
|
int2 dstTexCoord = int2(GlobalInvocationID.xy);
|
|
int2 srcTexCoord = dstTexCoord;
|
|
if ((uniforms.dstTextureFlipY == 1u)) {
|
|
srcTexCoord.y = as_type<int>((as_type<uint>(as_type<int>((as_type<uint>(size.y) - as_type<uint>(dstTexCoord.y)))) - as_type<uint>(1)));
|
|
}
|
|
float4 srcColor = tint_symbol_1.read(uint2(srcTexCoord), 0);
|
|
float4 dstColor = tint_symbol_2.read(uint2(dstTexCoord), 0);
|
|
bool success = true;
|
|
uint4 srcColorBits = 0u;
|
|
uint4 dstColorBits = uint4(dstColor);
|
|
for(uint i = 0u; (i < uniforms.channelCount); i = (i + 1u)) {
|
|
srcColorBits[i] = ConvertToFp16FloatValue(srcColor[i]);
|
|
success = (success && (srcColorBits[i] == dstColorBits[i]));
|
|
}
|
|
uint outputIndex = ((GlobalInvocationID.y * uint(size.x)) + GlobalInvocationID.x);
|
|
if (success) {
|
|
output.result[outputIndex] = uint(1);
|
|
} else {
|
|
output.result[outputIndex] = uint(0);
|
|
}
|
|
}
|
|
|
|
kernel void tint_symbol(texture2d<float, access::sample> tint_symbol_3 [[texture(0)]], texture2d<float, access::sample> tint_symbol_4 [[texture(1)]], uint3 GlobalInvocationID [[thread_position_in_grid]], constant Uniforms& uniforms [[buffer(3)]], device OutputBuf& output [[buffer(2)]]) {
|
|
tint_symbol_inner(uniforms, output, GlobalInvocationID, tint_symbol_3, tint_symbol_4);
|
|
return;
|
|
}
|
|
|