Skip to content

GitLab

Unverified Commit c23a7072 authored by Elfie Guo's avatar Elfie Guo Committed by GitHub
Browse files

Upgrade CUTLASS 4.0 (#6336) 


Co-authored-by: default avatarzhyncs <me@zhyncs.com>
parent e07a6977
Hide whitespace changes
Inline Side-by-side
Showing with 2 additions and 2 deletions
sgl-kernel/CMakeLists.txt
View file @ c23a7072
......@@ -45,7 +45,7 @@ include(FetchContent)
FetchContent_Declare(
repo-cutlass
GIT_REPOSITORY https://github.com/NVIDIA/cutlass
GIT_TAG ad7b2f5e84fcfa124cb02b91d5bd26d238c0459e
GIT_TAG f115c3f85467d5d9619119d1dbeb9c03c3d73864
GIT_SHALLOW OFF
)
FetchContent_Populate(repo-cutlass)
......
sgl-kernel/csrc/gemm/fp8_blockwise_gemm_kernel.cu 100644 → 100755
View file @ c23a7072
......@@ -140,7 +140,7 @@ void launch_sm90_fp8_blockwise_scaled_mm(
LayoutSFB layout_sfb = ScaleConfig::tile_atom_to_shape_SFB(make_shape(m, n, k, 1));
typename GemmKernel::MainloopArguments mainloop_args{
a_ptr, stride_a, b_ptr, stride_b, 4, a_s_ptr, layout_sfa, b_s_ptr, layout_sfb};
a_ptr, stride_a, b_ptr, stride_b, a_s_ptr, layout_sfa, b_s_ptr, layout_sfb};
typename GemmKernel::EpilogueArguments epilogue_args{{}, nullptr, stride_d, o_ptr, stride_d};
typename Gemm::Arguments args = {
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment