xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-01-27 12:05:59 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Misc][ROCm] Exclude cutlass_mla_decode for ROCm build (#17289)

Browse Source 
Signed-off-by: Tianyuan Wu <Tianyuan.Wu@amd.com>
This commit is contained in:
TY-AMD 2025-04-30 01:26:42 +08:00 committed by GitHub
parent d3cf61b89b
commit 06ffc7e1d3
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 7 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

14
csrc/torch_bindings.cpp
View File

@ -130,13 +130,6 @@ TORCH_LIBRARY_EXPAND(TORCH_EXTENSION_NAME, ops) {
") -> ()");
ops.impl("advance_step_flashinfer", torch::kCUDA, &advance_step_flashinfer);
// Compute MLA decode using cutlass.
ops.def(
"cutlass_mla_decode(Tensor! out, Tensor q_nope, Tensor q_pe,"
" Tensor kv_c_and_k_pe_cache, Tensor seq_lens,"
" Tensor page_table, float scale) -> ()");
ops.impl("cutlass_mla_decode", torch::kCUDA, &cutlass_mla_decode);
// Layernorm
// Apply Root Mean Square (RMS) Normalization to the input tensor.
ops.def(
@ -450,6 +443,13 @@ TORCH_LIBRARY_EXPAND(TORCH_EXTENSION_NAME, ops) {
ops.def("cutlass_sparse_compress(Tensor a) -> Tensor[]");
ops.impl("cutlass_sparse_compress", &cutlass_sparse_compress);
// CUTLASS MLA decode
ops.def(
"cutlass_mla_decode(Tensor! out, Tensor q_nope, Tensor q_pe,"
" Tensor kv_c_and_k_pe_cache, Tensor seq_lens,"
" Tensor page_table, float scale) -> ()");
ops.impl("cutlass_mla_decode", torch::kCUDA, &cutlass_mla_decode);
// Mamba selective scan kernel
ops.def(
"selective_scan_fwd(Tensor! u, Tensor! delta,"