d1479f871a
This avoids a lot of huge code changes. Arrays generally cannot be copied in and out of buffers, at least no compiler frontend seems to do it. Also avoids a lot of issues surrounding packed vectors and matrices.
36 lines
746 B
Plaintext
36 lines
746 B
Plaintext
#pragma clang diagnostic ignored "-Wunused-variable"
|
|
|
|
#include <metal_stdlib>
|
|
#include <simd/simd.h>
|
|
#include <metal_atomic>
|
|
|
|
using namespace metal;
|
|
|
|
struct SSBO
|
|
{
|
|
float in_data[1];
|
|
};
|
|
|
|
struct SSBO2
|
|
{
|
|
float out_data[1];
|
|
};
|
|
|
|
struct SSBO3
|
|
{
|
|
uint count;
|
|
};
|
|
|
|
constant uint3 gl_WorkGroupSize [[maybe_unused]] = uint3(4u, 1u, 1u);
|
|
|
|
kernel void main0(const device SSBO& _22 [[buffer(0)]], device SSBO2& _38 [[buffer(1)]], device SSBO3& _41 [[buffer(2)]], uint3 gl_GlobalInvocationID [[thread_position_in_grid]])
|
|
{
|
|
float _28 = _22.in_data[gl_GlobalInvocationID.x];
|
|
if (_28 > 12.0)
|
|
{
|
|
uint _45 = atomic_fetch_add_explicit((device atomic_uint*)&_41.count, 1u, memory_order_relaxed);
|
|
_38.out_data[_45] = _28;
|
|
}
|
|
}
|
|
|