what is rasterization in Cutlass? what is the relationship between swizzle and rasterization? Any documents?
Here is a reference. That whole page may be of interest.
You may also wish to refer to the resources indicated at the top of that github page, i.e.:
CUTLASS implements the hierarchically blocked structure described in CUTLASS: Fast Linear Algebra in CUDA C++ and the CUTLASS GTC2018 talk.