EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

fix: remove cublas_grouped_gemm (#5307)

Browse Source
This commit is contained in:
Yineng Zhang
2025-04-11 16:22:37 -07:00
committed by GitHub
parent 034c5256cc
commit 136b8e6afb
8 changed files with 0 additions and 508 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
sgl-kernel/csrc/common_extension.cc
View File

@@ -112,11 +112,6 @@ TORCH_LIBRARY_FRAGMENT(sgl_kernel, m) {
m.def("sgl_per_token_quant_fp8(Tensor input, Tensor output_q, Tensor output_s) -> ()");
m.impl("sgl_per_token_quant_fp8", torch::kCUDA, &sgl_per_token_quant_fp8);
m.def(
"cublas_grouped_gemm(Tensor[] inputs, Tensor[] weights, Tensor[] outputs,"
" ScalarType out_dtype, int cublas_handle, int cuda_stream) -> ()");
m.impl("cublas_grouped_gemm", torch::kCUDA, &cublas_grouped_gemm);
m.def(
"cutlass_scaled_fp4_mm(Tensor! out, Tensor a, Tensor b,"
" Tensor block_scale_a, Tensor block_scale_b,"