SPIRV-Cross/reference/shaders-msl/frag/array-of-texture-swizzle.msl2.swizzle.frag
Hans-Kristian Arntzen 5b952d2cbf MSL: Rethink how opaque descriptors are passed to leaf functions.
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.
2022-01-18 14:40:52 +01:00

99 lines
2.8 KiB
JavaScript

#pragma clang diagnostic ignored "-Wmissing-prototypes"
#include <metal_stdlib>
#include <simd/simd.h>
using namespace metal;
template<typename T> struct spvRemoveReference { typedef T type; };
template<typename T> struct spvRemoveReference<thread T&> { typedef T type; };
template<typename T> struct spvRemoveReference<thread T&&> { typedef T type; };
template<typename T> inline constexpr thread T&& spvForward(thread typename spvRemoveReference<T>::type& x)
{
return static_cast<thread T&&>(x);
}
template<typename T> inline constexpr thread T&& spvForward(thread typename spvRemoveReference<T>::type&& x)
{
return static_cast<thread T&&>(x);
}
enum class spvSwizzle : uint
{
none = 0,
zero,
one,
red,
green,
blue,
alpha
};
template<typename T>
inline T spvGetSwizzle(vec<T, 4> x, T c, spvSwizzle s)
{
switch (s)
{
case spvSwizzle::none:
return c;
case spvSwizzle::zero:
return 0;
case spvSwizzle::one:
return 1;
case spvSwizzle::red:
return x.r;
case spvSwizzle::green:
return x.g;
case spvSwizzle::blue:
return x.b;
case spvSwizzle::alpha:
return x.a;
}
}
// Wrapper function that swizzles texture samples and fetches.
template<typename T>
inline vec<T, 4> spvTextureSwizzle(vec<T, 4> x, uint s)
{
if (!s)
return x;
return vec<T, 4>(spvGetSwizzle(x, x.r, spvSwizzle((s >> 0) & 0xFF)), spvGetSwizzle(x, x.g, spvSwizzle((s >> 8) & 0xFF)), spvGetSwizzle(x, x.b, spvSwizzle((s >> 16) & 0xFF)), spvGetSwizzle(x, x.a, spvSwizzle((s >> 24) & 0xFF)));
}
template<typename T>
inline T spvTextureSwizzle(T x, uint s)
{
return spvTextureSwizzle(vec<T, 4>(x, 0, 0, 1), s).x;
}
struct main0_out
{
float4 FragColor [[color(0)]];
};
struct main0_in
{
float2 vUV [[user(locn0)]];
};
static inline __attribute__((always_inline))
float4 sample_in_func(thread const array<texture2d<float>, 4>& uSampler, thread const array<sampler, 4>& uSamplerSmplr, constant uint* uSamplerSwzl, thread float2& vUV)
{
return spvTextureSwizzle(uSampler[2].sample(uSamplerSmplr[2], vUV), uSamplerSwzl[2]);
}
static inline __attribute__((always_inline))
float4 sample_single_in_func(texture2d<float> s, sampler sSmplr, constant uint& sSwzl, thread float2& vUV)
{
return spvTextureSwizzle(s.sample(sSmplr, vUV), sSwzl);
}
fragment main0_out main0(main0_in in [[stage_in]], constant uint* spvSwizzleConstants [[buffer(30)]], array<texture2d<float>, 4> uSampler [[texture(0)]], array<sampler, 4> uSamplerSmplr [[sampler(0)]])
{
main0_out out = {};
constant uint* uSamplerSwzl = &spvSwizzleConstants[0];
out.FragColor = sample_in_func(uSampler, uSamplerSmplr, uSamplerSwzl, in.vUV);
out.FragColor += sample_single_in_func(uSampler[1], uSamplerSmplr[1], uSamplerSwzl[1], in.vUV);
return out;
}