    162     0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
      8 ptxas info    : Used 51 registers, used 1 barriers, 128 bytes smem, 580 bytes cmem[0], 8 bytes cmem[2]
      8 ptxas info    : Used 46 registers, used 1 barriers, 128 bytes smem, 588 bytes cmem[0], 8 bytes cmem[2]
      8 ptxas info    : Used 46 registers, used 1 barriers, 128 bytes smem, 580 bytes cmem[0], 8 bytes cmem[2]
      8 ptxas info    : Used 41 registers, used 1 barriers, 128 bytes smem, 588 bytes cmem[0], 8 bytes cmem[2]
      6 ptxas info    : Used 62 registers, used 1 barriers, 128 bytes smem, 588 bytes cmem[0], 8 bytes cmem[2]
