Swipe the game board to move all tiles left, right, up, or down. When two tiles with the same number touch, they merge into one with a double number, i.e. 2+2 merge into 4, 4+4 merge into 8 and so on.
`gl.nvidia.blackwell.tma.async_scatter` functions respectively. TMA gather and scatter operations only support 2D tensor descriptors, where the first dimension of the block shape must be 1. Gather ...