How to Optimize a CUDA Matmul Kernel for CuBLAS-Like Performance: A Worklog

Re:How to Optimize a CUDA Matmul Kernel for CuBLAS-Like Perf... by lights0123 on Friday July 26, 2024 @10:22PM (#41084043)

Since 2016, you can use Eigen in CUDA C++ code and it just works: https://eigen.tuxfamily.org/dox/TopicCUDA.html

[ Parent | on HN ]

Re:How to Optimize a CUDA Matmul Kernel for CuBLAS-Like Perf... by bee_rider on Saturday July 27, 2024 @09:07AM (#41086335)

Oh wow, that is cool as heck.

[ Parent | on HN ]