[WideEP] Remove pplx all2all backend (#33724)

Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
Tyler Michael Smith
2026-02-26 14:30:10 -08:00
committed by GitHub
co-authored by Claude Opus 4.6
parent 0f2f24c8b2
commit eb19955c37
39 changed files with 107 additions and 2069 deletions
+4 -4
View File
@@ -489,19 +489,19 @@ TORCH_LIBRARY_EXPAND(TORCH_EXTENSION_NAME, ops) {
&get_cutlass_moe_mm_problem_sizes_from_expert_offsets);
// A function that computes data required to run fused MoE with w8a8 grouped
// GEMM and PPLX. It takes expert_num_tokens and non_zero_expert_idxs
// GEMM in batched expert format. It takes expert_num_tokens
// as an input, and computes expert_offsets (token start indices of each
// expert). In addition to this, it computes problem sizes for each expert's
// multiplication used by the two mms called from fused MoE operation.
ops.def(
"get_cutlass_pplx_moe_mm_data(Tensor! expert_offsets, "
"get_cutlass_batched_moe_mm_data(Tensor! expert_offsets, "
" Tensor! problem_sizes1, "
" Tensor! problem_sizes2, "
" Tensor expert_num_tokens, "
" int num_local_experts, int padded_m, "
" int n, int k) -> ()");
ops.impl("get_cutlass_pplx_moe_mm_data", torch::kCUDA,
&get_cutlass_pplx_moe_mm_data);
ops.impl("get_cutlass_batched_moe_mm_data", torch::kCUDA,
&get_cutlass_batched_moe_mm_data);
// Check if cutlass scaled_mm supports block quantization (used by DeepSeekV3)
ops.def(