Has anyone ever tried blocking on 3D matrices using CUDA? If someone has done so, it’ll be great if you could share the source…I’m facing a lotta trouble in coding this part and need to do blockings (either cube or pencil shaped) on huge 3d matrices [1k1k1k]? Anyone?
Thanks in advance,
PS : Since I’m not really a computer guy, help in the form of links / papers for 3d blocking (especially the math part!) would also be great!