[Refactor] Remove dead cutlass mxfp8 code (#44681)

Signed-off-by: yewentao256 <zhyanwentao@126.com>
Co-authored-by: Shengqi Chen <harry-chen@outlook.com>
This commit is contained in:
Wentao Ye
2026-06-17 21:18:25 -07:00
committed by GitHub
co-authored by Shengqi Chen
parent f428718ffe
commit b4c80ec0fd
10 changed files with 0 additions and 1369 deletions
-16
View File
@@ -308,22 +308,6 @@ STABLE_TORCH_LIBRARY_FRAGMENT(_C, ops) {
"awq_dequantize(Tensor _kernel, Tensor _scaling_factors, "
"Tensor _zeros, SymInt split_k_iters, int thx, int thy) -> Tensor");
// Expert-specialization mxfp8 blockscaled grouped quantization (SM100+).
ops.def(
"mxfp8_experts_quant("
" Tensor input, Tensor problem_sizes, Tensor expert_offsets,"
" Tensor blockscale_offsets, Tensor! quant_output, Tensor! scale_factor)"
" -> ()");
// conditionally compiled so impl registration is in source file
// Expert-specialization mxfp8 blockscaled grouped GEMM (SM100+).
ops.def(
"cutlass_mxfp8_grouped_mm("
" Tensor a, Tensor b, Tensor sfa, Tensor sfb, Tensor! out,"
" Tensor problem_sizes, Tensor expert_offsets, Tensor blockscale_offsets)"
" -> ()");
// conditionally compiled so impl registration is in source file
// DeepSeek V3 fused A GEMM (SM 9.0+, bf16 only, 1-16 tokens).
// conditionally compiled so impl registration is in source file
ops.def(