CUTLASS
CUDA Templates for Linear Algebra Subroutines and Solvers

gemm → threadblock Relation

File in include/cutlass/gemmIncludes file in include/cutlass/epilogue/threadblock
kernel / default_gemm.hdefault_epilogue_simt.h
kernel / default_gemm.hdefault_epilogue_tensor_op.h
kernel / default_gemm.hdefault_epilogue_volta_tensor_op.h
kernel / default_gemm.hepilogue.h