minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 13 Packages Projects Releases Wiki Activity

[CPU] remove RECORD_FUNCTION (#23528)

Browse Source
This commit is contained in:
Ma Mingfei
2026-04-24 09:18:29 +08:00
committed by GitHub
parent 87e50f20f6
commit 23e4d381f0
21 changed files with 0 additions and 106 deletions
Download Patch File Download Diff File Expand all files Collapse all files
sgl-kernel/csrc/cpu/gemm_int8.cpp
-6
View File
@@ -380,8 +380,6 @@ INSTANTIATE_TINYGEMM_TEMPLATE(at::BFloat16);
INSTANTIATE_TINYGEMM_TEMPLATE(at::Half);
std::tuple<at::Tensor, at::Tensor> per_token_quant_int8_cpu(at::Tensor& A) {
RECORD_FUNCTION("sgl-kernel::per_token_quant_int8_cpu", std::vector<c10::IValue>({A}));
CHECK_LAST_DIM_CONTIGUOUS_INPUT(A);
CHECK_DIM(2, A);
@@ -427,8 +425,6 @@ at::Tensor int8_scaled_mm_cpu(
const std::optional<at::Tensor>& bias,
at::ScalarType out_dtype,
bool is_vnni) {
RECORD_FUNCTION("sgl-kernel::int8_scaled_mm_cpu", std::vector<c10::IValue>({mat1, mat2, scales1, scales2, bias}));
auto packed_w = is_vnni ? mat2 : convert_weight_packed(mat2);
CHECK_INPUT(mat1);
@@ -485,8 +481,6 @@ at::Tensor int8_scaled_mm_with_quant(
const std::optional<at::Tensor>& bias,
at::ScalarType out_dtype,
bool is_vnni) {
RECORD_FUNCTION("sgl-kernel::int8_scaled_mm_cpu", std::vector<c10::IValue>({mat1, mat2, scales2, bias}));
auto packed_w = is_vnni ? mat2 : convert_weight_packed(mat2);
CHECK_LAST_DIM_CONTIGUOUS_INPUT(mat1);