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.
34 lines
836 B
Plaintext
34 lines
836 B
Plaintext
#pragma clang diagnostic ignored "-Wmissing-prototypes"
|
|
|
|
#include <metal_stdlib>
|
|
#include <simd/simd.h>
|
|
|
|
using namespace metal;
|
|
|
|
struct ssbo
|
|
{
|
|
uint _data[1];
|
|
};
|
|
|
|
static inline __attribute__((always_inline))
|
|
void Load(thread const uint& size, const device ssbo& ssbo_1)
|
|
{
|
|
int byteAddrTemp = int(size >> uint(2));
|
|
uint4 data = uint4(ssbo_1._data[byteAddrTemp], ssbo_1._data[byteAddrTemp + 1], ssbo_1._data[byteAddrTemp + 2], ssbo_1._data[byteAddrTemp + 3]);
|
|
}
|
|
|
|
static inline __attribute__((always_inline))
|
|
void _main(thread const uint3& id, const device ssbo& ssbo_1)
|
|
{
|
|
uint param = 4u;
|
|
Load(param, ssbo_1);
|
|
}
|
|
|
|
kernel void main0(const device ssbo& ssbo_1 [[buffer(0)]], uint3 gl_GlobalInvocationID [[thread_position_in_grid]])
|
|
{
|
|
uint3 id = gl_GlobalInvocationID;
|
|
uint3 param = id;
|
|
_main(param, ssbo_1);
|
|
}
|
|
|