5b952d2cbf
We were passing arrays by value which the compiler fails to optimize, causing abyssal performance. To fix this, we need to consider that descriptors can be in constant or const device address spaces. Also, lone descriptors are passed by value, so we explicitly remove address space qualifiers. One failure case is when shader passes a texture/sampler array as an argument. It's all UniformConstant in SPIR-V, but in MSL it might be thread, const device or constant, so that won't work ... Global variable use works fine though, and that should cover 99.9999999% of use cases.
32 lines
851 B
GLSL
32 lines
851 B
GLSL
#pragma clang diagnostic ignored "-Wmissing-prototypes"
|
|
|
|
#include <metal_stdlib>
|
|
#include <simd/simd.h>
|
|
|
|
using namespace metal;
|
|
|
|
struct main0_out
|
|
{
|
|
float FragColor [[color(0)]];
|
|
};
|
|
|
|
static inline __attribute__((always_inline))
|
|
float sample_depth_from_function(depth2d<float> uT, sampler uS)
|
|
{
|
|
return uT.sample_compare(uS, float3(0.5).xy, 0.5);
|
|
}
|
|
|
|
static inline __attribute__((always_inline))
|
|
float sample_color_from_function(texture2d<float> uT, sampler uS)
|
|
{
|
|
return uT.sample(uS, float2(0.5)).x;
|
|
}
|
|
|
|
fragment main0_out main0(depth2d<float> uDepth [[texture(0)]], texture2d<float> uColor [[texture(1)]], sampler uSamplerShadow [[sampler(0)]], sampler uSampler [[sampler(1)]])
|
|
{
|
|
main0_out out = {};
|
|
out.FragColor = sample_depth_from_function(uDepth, uSamplerShadow) + sample_color_from_function(uColor, uSampler);
|
|
return out;
|
|
}
|
|
|