SPIRV-Cross/reference/opt/shaders-msl/comp/force-recompile-hooks.swizzle.comp

74 lines
2.1 KiB
Plaintext
Raw Normal View History

#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; };
2019-09-24 22:13:04 +00:00
template<typename T> inline constexpr thread T&& spvForward(thread typename spvRemoveReference<T>::type& x)
{
return static_cast<thread T&&>(x);
}
2019-09-24 22:13:04 +00:00
template<typename T> inline constexpr thread T&& spvForward(thread typename spvRemoveReference<T>::type&& x)
{
return static_cast<thread T&&>(x);
}
MSL: Add support for sampler Y'CbCr conversion. This change introduces functions and in one case, a class, to support the `VK_KHR_sampler_ycbcr_conversion` extension. Except in the case of GBGR8 and BGRG8 formats, for which Metal natively supports implicit chroma reconstruction, we're on our own here. We have to do everything ourselves. Much of the complexity comes from the need to support multiple planes, which must now be passed to functions that use the corresponding combined image-samplers. The rest is from the actual Y'CbCr conversion itself, which requires additional post-processing of the sample retrieved from the image. Passing sampled images to a function was a particular problem. To support this, I've added a new class which is emitted to MSL shaders that pass sampled images with Y'CbCr conversions attached around. It can handle sampled images with or without Y'CbCr conversion. This is an awful abomination that should not exist, but I'm worried that there's some shader out there which does this. This support requires Metal 2.0 to work properly, because it uses default-constructed texture objects, which were only added in MSL 2. I'm not even going to get into arrays of combined image-samplers--that's a whole other can of worms. They are deliberately unsupported in this change. I've taken the liberty of refactoring the support for texture swizzling while I'm at it. It's now treated as a post-processing step similar to Y'CbCr conversion. I'd like to think this is cleaner than having everything in `to_function_name()`/`to_function_args()`. It still looks really hairy, though. I did, however, get rid of the explicit type arguments to `spvGatherSwizzle()`/`spvGatherCompareSwizzle()`. Update the C API. In addition to supporting this new functionality, add some compiler options that I added in previous changes, but for which I neglected to update the C API.
2019-08-02 20:11:19 +00:00
enum class spvSwizzle : uint
{
none = 0,
zero,
one,
red,
green,
blue,
alpha
};
template<typename T>
2019-09-23 22:05:04 +00:00
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>
2019-09-23 22:05:04 +00:00
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>
2019-09-23 22:05:04 +00:00
inline T spvTextureSwizzle(T x, uint s)
{
return spvTextureSwizzle(vec<T, 4>(x, 0, 0, 1), s).x;
}
2019-09-24 22:13:04 +00:00
kernel void main0(constant uint* spvSwizzleConstants [[buffer(30)]], texture2d<float> foo [[texture(0)]], texture2d<float, access::write> bar [[texture(1)]], sampler fooSmplr [[sampler(0)]])
{
2019-09-24 22:13:04 +00:00
constant uint& fooSwzl = spvSwizzleConstants[0];
bar.write(spvTextureSwizzle(foo.sample(fooSmplr, float2(1.0), level(0.0)), fooSwzl), uint2(int2(0)));
}