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

[torch.compile] Add torch inductor pass for fusing silu_and_mul with subsequent scaled_fp8_quant operations (#10867)

Browse Source 
Signed-off-by: Sage Moore <sage@neuralmagic.com>
This commit is contained in:
Sage Moore
2025-05-01 07:59:28 -07:00
committed by GitHub
parent 28566d73b3
commit 460a2b1100
11 changed files with 406 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
csrc/torch_bindings.cpp
View File

@ -81,9 +81,13 @@ TORCH_LIBRARY_EXPAND(TORCH_EXTENSION_NAME, ops) {
// Activation ops
// Activation function used in SwiGLU.
ops.def("silu_and_mul(Tensor! out, Tensor input) -> ()");
ops.def("silu_and_mul(Tensor! result, Tensor input) -> ()");
ops.impl("silu_and_mul", torch::kCUDA, &silu_and_mul);
ops.def(
"silu_and_mul_quant(Tensor! result, Tensor input, Tensor scale) -> ()");
ops.impl("silu_and_mul_quant", torch::kCUDA, &silu_and_mul_quant);
ops.def("mul_and_silu(Tensor! out, Tensor input) -> ()");
ops.impl("mul_and_silu", torch::kCUDA, &mul_and_silu);