Commit d6a856f5 authored by zhuwenwen's avatar zhuwenwen
Browse files

skip silu_and_mul_quant

parent e87de76c
...@@ -260,8 +260,8 @@ void rotary_embedding_tgi( ...@@ -260,8 +260,8 @@ void rotary_embedding_tgi(
void silu_and_mul(torch::Tensor& out, torch::Tensor& input); void silu_and_mul(torch::Tensor& out, torch::Tensor& input);
void silu_and_mul_quant(torch::Tensor& out, torch::Tensor& input, // void silu_and_mul_quant(torch::Tensor& out, torch::Tensor& input,
torch::Tensor& scale); // torch::Tensor& scale);
void mul_and_silu(torch::Tensor& out, torch::Tensor& input); void mul_and_silu(torch::Tensor& out, torch::Tensor& input);
......
...@@ -259,9 +259,9 @@ TORCH_LIBRARY_EXPAND(TORCH_EXTENSION_NAME, ops) { ...@@ -259,9 +259,9 @@ TORCH_LIBRARY_EXPAND(TORCH_EXTENSION_NAME, ops) {
ops.def("silu_and_mul(Tensor! result, Tensor input) -> ()"); ops.def("silu_and_mul(Tensor! result, Tensor input) -> ()");
ops.impl("silu_and_mul", torch::kCUDA, &silu_and_mul); ops.impl("silu_and_mul", torch::kCUDA, &silu_and_mul);
ops.def( // ops.def(
"silu_and_mul_quant(Tensor! result, Tensor input, Tensor scale) -> ()"); // "silu_and_mul_quant(Tensor! result, Tensor input, Tensor scale) -> ()");
ops.impl("silu_and_mul_quant", torch::kCUDA, &silu_and_mul_quant); // ops.impl("silu_and_mul_quant", torch::kCUDA, &silu_and_mul_quant);
ops.def("mul_and_silu(Tensor! out, Tensor input) -> ()"); ops.def("mul_and_silu(Tensor! out, Tensor input) -> ()");
ops.impl("mul_and_silu", torch::kCUDA, &mul_and_silu); ops.impl("mul_and_silu", torch::kCUDA, &mul_and_silu);
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment