111 lines
3.1 KiB
Plaintext
111 lines
3.1 KiB
Plaintext
#pragma clang diagnostic ignored "-Wmissing-braces"
|
|
#pragma clang diagnostic ignored "-Wunused-variable"
|
|
|
|
#include <metal_stdlib>
|
|
#include <simd/simd.h>
|
|
|
|
template <typename T, size_t Num>
|
|
struct unsafe_array
|
|
{
|
|
T __Elements[Num ? Num : 1];
|
|
|
|
constexpr size_t size() const thread { return Num; }
|
|
constexpr size_t max_size() const thread { return Num; }
|
|
constexpr bool empty() const thread { return Num == 0; }
|
|
|
|
constexpr size_t size() const device { return Num; }
|
|
constexpr size_t max_size() const device { return Num; }
|
|
constexpr bool empty() const device { return Num == 0; }
|
|
|
|
constexpr size_t size() const constant { return Num; }
|
|
constexpr size_t max_size() const constant { return Num; }
|
|
constexpr bool empty() const constant { return Num == 0; }
|
|
|
|
constexpr size_t size() const threadgroup { return Num; }
|
|
constexpr size_t max_size() const threadgroup { return Num; }
|
|
constexpr bool empty() const threadgroup { return Num == 0; }
|
|
|
|
thread T &operator[](size_t pos) thread
|
|
{
|
|
return __Elements[pos];
|
|
}
|
|
constexpr const thread T &operator[](size_t pos) const thread
|
|
{
|
|
return __Elements[pos];
|
|
}
|
|
|
|
device T &operator[](size_t pos) device
|
|
{
|
|
return __Elements[pos];
|
|
}
|
|
constexpr const device T &operator[](size_t pos) const device
|
|
{
|
|
return __Elements[pos];
|
|
}
|
|
|
|
constexpr const constant T &operator[](size_t pos) const constant
|
|
{
|
|
return __Elements[pos];
|
|
}
|
|
|
|
threadgroup T &operator[](size_t pos) threadgroup
|
|
{
|
|
return __Elements[pos];
|
|
}
|
|
constexpr const threadgroup T &operator[](size_t pos) const threadgroup
|
|
{
|
|
return __Elements[pos];
|
|
}
|
|
};
|
|
|
|
using namespace metal;
|
|
|
|
struct Sub
|
|
{
|
|
unsafe_array<float,2> f;
|
|
unsafe_array<float2,2> f2;
|
|
unsafe_array<float3,2> f3;
|
|
unsafe_array<float4,2> f4;
|
|
};
|
|
|
|
struct Sub_1
|
|
{
|
|
unsafe_array<float4,2> f;
|
|
unsafe_array<float4,2> f2;
|
|
unsafe_array<float3,2> f3;
|
|
unsafe_array<float4,2> f4;
|
|
};
|
|
|
|
struct SSBO
|
|
{
|
|
unsafe_array<Sub_1,2> sub;
|
|
};
|
|
|
|
kernel void main0(device SSBO& _27 [[buffer(0)]], uint3 gl_WorkGroupID [[threadgroup_position_in_grid]], uint3 gl_GlobalInvocationID [[thread_position_in_grid]])
|
|
{
|
|
Sub foo;
|
|
foo.f[0] = _27.sub[gl_WorkGroupID.x].f[0].x;
|
|
foo.f[1] = _27.sub[gl_WorkGroupID.x].f[1].x;
|
|
foo.f2[0] = _27.sub[gl_WorkGroupID.x].f2[0].xy;
|
|
foo.f2[1] = _27.sub[gl_WorkGroupID.x].f2[1].xy;
|
|
foo.f3[0] = _27.sub[gl_WorkGroupID.x].f3[0];
|
|
foo.f3[1] = _27.sub[gl_WorkGroupID.x].f3[1];
|
|
foo.f4[0] = _27.sub[gl_WorkGroupID.x].f4[0];
|
|
foo.f4[1] = _27.sub[gl_WorkGroupID.x].f4[1];
|
|
foo.f[gl_GlobalInvocationID.x] += 1.0;
|
|
foo.f2[gl_GlobalInvocationID.x] += float2(2.0);
|
|
foo.f3[gl_GlobalInvocationID.x] += float3(3.0);
|
|
foo.f4[gl_GlobalInvocationID.x] += float4(4.0);
|
|
_27.sub[gl_WorkGroupID.x].f[0].x = foo.f[0];
|
|
_27.sub[gl_WorkGroupID.x].f[1].x = foo.f[1];
|
|
_27.sub[gl_WorkGroupID.x].f2[0].xy = foo.f2[0];
|
|
_27.sub[gl_WorkGroupID.x].f2[1].xy = foo.f2[1];
|
|
_27.sub[gl_WorkGroupID.x].f3[0] = foo.f3[0];
|
|
_27.sub[gl_WorkGroupID.x].f3[1] = foo.f3[1];
|
|
_27.sub[gl_WorkGroupID.x].f4[0] = foo.f4[0];
|
|
_27.sub[gl_WorkGroupID.x].f4[1] = foo.f4[1];
|
|
_27.sub[0].f[0].x = float(_27.sub[0].f[0].x) + 5.0;
|
|
_27.sub[0].f2[1].xy += float2(5.0);
|
|
}
|
|
|