search

TiledTensor / TiledCUDA

TiledCUDA is a highly efficient kernel template library designed to elevate CUDA C’s level of abstraction for processing tiles.
MIT License
157 stars 10 forks source link

fix(cell): bug fix for global to shared memory data loader. #126

Closed haruhi55 closed 3 months ago

haruhi55 commented 3 months ago
  1. bug fixed for global to shared data loader.
  2. add more unittest for fused two GEMMs.