21d38f74ce
Fix reversed coordinates: `y` should be used to calculate the row address. Align row address to the row stride. I've made the row alignment a function constant; this makes it possible to override it at pipeline compile time. Honestly, I don't know how this worked at all for Epic. It definitely didn't work in the CTS prior to this.
29 lines
1.4 KiB
Plaintext
29 lines
1.4 KiB
Plaintext
#pragma clang diagnostic ignored "-Wmissing-prototypes"
|
|
#pragma clang diagnostic ignored "-Wunused-variable"
|
|
|
|
#include <metal_stdlib>
|
|
#include <simd/simd.h>
|
|
#include <metal_atomic>
|
|
|
|
using namespace metal;
|
|
|
|
struct SSBO
|
|
{
|
|
float4 outdata;
|
|
};
|
|
|
|
constant uint3 gl_WorkGroupSize [[maybe_unused]] = uint3(1u);
|
|
|
|
// The required alignment of a linear texture of R32Uint format.
|
|
constant uint spvLinearTextureAlignmentOverride [[function_constant(65535)]];
|
|
constant uint spvLinearTextureAlignment = is_function_constant_defined(spvLinearTextureAlignmentOverride) ? spvLinearTextureAlignmentOverride : 4;
|
|
// Returns buffer coords corresponding to 2D texture coords for emulating 2D texture atomics
|
|
#define spvImage2DAtomicCoord(tc, tex) (((((tex).get_width() + spvLinearTextureAlignment / 4 - 1) & ~( spvLinearTextureAlignment / 4 - 1)) * (tc).y) + (tc).x)
|
|
|
|
kernel void main0(device SSBO& _31 [[buffer(1)]], texture2d<uint> uImage [[texture(0)]], device atomic_uint* uImage_atomic [[buffer(0)]], texture2d<float> uTexture [[texture(1)]], sampler uTextureSmplr [[sampler(0)]], uint3 gl_GlobalInvocationID [[thread_position_in_grid]])
|
|
{
|
|
uint _26 = atomic_fetch_add_explicit((device atomic_uint*)&uImage_atomic[spvImage2DAtomicCoord(int2(gl_GlobalInvocationID.xy), uImage)], 10u, memory_order_relaxed);
|
|
_31.outdata = uTexture.sample(uTextureSmplr, float2(gl_GlobalInvocationID.xy), level(0.0)) + float4(float(_26));
|
|
}
|
|
|