Quantization and 3D convolution

[tkonolige]

I believe this line is the issue as it occurs before threadIdx.z is defined.

@OValery16

allocate(compute, int32, [(((floordiv(((threadIdx.z: int32*2) + 1), 4)*32) + 32) - (floordiv(threadIdx.z, 2)*32))]);

However, I cannot reproduce this issue with the script you’ve given me. Are you up to date on master? I used c7c39a492a51736dbd780fa12021fee6097bfe88.