biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[Kernel] Integrate CUTLASS MoE kernel with PPLX (#18762)

Browse Source 
Signed-off-by: ElizaWszola <ewszola@redhat.com>
Signed-off-by: Tyler Michael Smith <tyler@neuralmagic.com>
Co-authored-by: Tyler Michael Smith <tyler@neuralmagic.com>
This commit is contained in:
ElizaWszola
2025-06-07 03:26:11 +02:00
committed by GitHub
parent 6e0cd10f72
commit 84166fee97
26 changed files with 918 additions and 409 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
csrc/quantization/cutlass_w8a8/moe/grouped_mm_c3x.cuh
View File

@@ -76,7 +76,8 @@ void cutlass_group_gemm_caller(
torch::Tensor const& b_tensors, torch::Tensor const& a_scales,
torch::Tensor const& b_scales, torch::Tensor const& expert_offsets,
torch::Tensor const& problem_sizes, torch::Tensor const& a_strides,
torch::Tensor const& b_strides, torch::Tensor const& c_strides) {
torch::Tensor const& b_strides, torch::Tensor const& c_strides,
bool per_act_token, bool per_out_ch) {
using ElementAB = typename Gemm::ElementAB;
using ElementD = typename Gemm::ElementD;
@@ -84,9 +85,6 @@ void cutlass_group_gemm_caller(
int k_size = a_tensors.size(1);
int n_size = out_tensors.size(1);
bool per_act_token = a_scales.numel() != 1;
bool per_out_ch = b_scales.numel() != num_experts;
auto stream = at::cuda::getCurrentCUDAStream(a_tensors.device().index());
auto options_int =