For the SCAN algorithm included with CUDA GPU SDK 4.0, the maximum array size is limited to 4*THREADBLOCK_SIZE * THREADBLOCK_SIZE, which is 262144. I try to raise the limit, but got wrong results. Can somebody explain why the maximum array size is limited?
For the SCAN algorithm included with CUDA GPU SDK 4.0, the maximum array size is limited to 4*THREADBLOCK_SIZE * THREADBLOCK_SIZE, which is 262144. I try to raise the limit, but got wrong results. Can somebody explain why the maximum array size is limited?