CUTLASS
CUDA Templates for Linear Algebra Subroutines and Solvers

kernel → threadblock Relation

File in include/cutlass/gemm/kernelIncludes file in include/cutlass/gemm/threadblock
default_gemm.hdefault_mma.h
default_gemm.hdefault_mma_core_simt.h
default_gemm.hdefault_mma_core_sm70.h
default_gemm.hdefault_mma_core_sm75.h
default_gemm.hgemm/threadblock/threadblock_swizzle.h
default_gemv.hdefault_gemv_core.h
default_gemv.hgemm/threadblock/threadblock_swizzle.h
default_gemv.hgemv.h