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

[Kernel] Add cuda kernel for gpt_oss activation (#22538)

Browse Source 
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
This commit is contained in:
Jee Jee Li
2025-08-15 08:21:29 +08:00
committed by GitHub
parent 39cd09dc86
commit 81f4b96481
8 changed files with 150 additions and 24 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
csrc/torch_bindings.cpp
View File

@ -130,6 +130,11 @@ TORCH_LIBRARY_EXPAND(TORCH_EXTENSION_NAME, ops) {
ops.def("fatrelu_and_mul(Tensor! out, Tensor input, float threshold) -> ()");
ops.impl("fatrelu_and_mul", torch::kCUDA, &fatrelu_and_mul);
ops.def(
"swigluoai_and_mul(Tensor! out, Tensor input, float alpha, float limit) "
"-> ()");
ops.impl("swigluoai_and_mul", torch::kCUDA, &swigluoai_and_mul);
// GELU implementation used in GPT-2.
ops.def("gelu_new(Tensor! out, Tensor input) -> ()");
ops.impl("gelu_new", torch::kCUDA, &gelu_new);