Roofline version with warp raked threadwise transfer.

b11ce4ba · Adam Osewski · 5c7e8b67 · b11ce4ba
Commit b11ce4ba authored Jun 15, 2023 by Adam Osewski
Expand all Hide whitespace changes
Inline Side-by-side

Showing with 996 additions and 0 deletions

include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_splitk_direct_c_write_out_roofline.hpp ...idwise_gemm_xdlops_splitk_direct_c_write_out_roofline.hpp +996 -0

No files found.
--- a/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_splitk_direct_c_write_out_roofline.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_gemm_xdlops_splitk_direct_c_write_out_roofline.hpp