vrr/kvcache-ai-ktransformers
2
0
Fork 0
mirror of https://github.com/kvcache-ai/ktransformers.git synced 2026-04-28 03:39:48 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 4

[build] prepare v0.6.1 SFT wheel packaging on main (#1945)
Some checks failed
Book-CI / test (push) Waiting to run
Details
Book-CI / test-1 (push) Waiting to run
Details
Book-CI / test-2 (push) Waiting to run
Details
Deploy / deploy (macos-latest) (push) Waiting to run
Details
Deploy / deploy (ubuntu-latest) (push) Waiting to run
Details
Deploy / deploy (windows-latest) (push) Waiting to run
Details
Release Fake Tag / publish (push) Has been cancelled
Details
Release to PyPI / Build & publish sglang-kt (push) Has been cancelled
Details
Release to PyPI / Build kt-kernel (Python 3.11) (push) Has been cancelled
Details
Release to PyPI / Build kt-kernel (Python 3.12) (push) Has been cancelled
Details
Release sglang-kt to PyPI / Build sglang-kt wheel (push) Has been cancelled
Details
Release to PyPI / Publish kt-kernel to PyPI (push) Has been cancelled
Details
Release sglang-kt to PyPI / Publish sglang-kt to PyPI (push) Has been cancelled
Details

Browse source 
* [build]: prepare 0.6.1 SFT wheel packaging on main

* [build]: finalize py311+ wheel packaging defaults
This commit is contained in:
Peilin Li 2026-04-24 12:08:38 +08:00 • committed by GitHub
parent 9544a8960d
commit 85308615b9
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
15 changed files with 305 additions and 35 deletions
Download patch file Download diff file Expand all files Collapse all files

10
kt-kernel/ext_bindings.cpp
View file

@ -12,7 +12,9 @@
#include <sys/wait.h>
#include <unistd.h>
#if defined(KTRANSFORMERS_ENABLE_CPPTRACE)
#include <cpptrace/cpptrace.hpp>
#endif
#include <csignal>
#include <cstddef>
#include <cstring>
@ -54,8 +56,8 @@ static const bool _is_plain_ = false;
#if defined(__x86_64__)
#include "operators/avx2/bf16-moe.hpp"
#include "operators/avx2/fp8-moe.hpp"
#include "operators/avx2/gptq_int4_avxvnni-moe.hpp"
#include "operators/avx2/gptq_int4-moe.hpp"
#include "operators/avx2/gptq_int4_avxvnni-moe.hpp"
#endif
#include <pybind11/stl.h> // std::vector/std::pair/std::string conversions
@ -74,7 +76,6 @@ static const bool _is_plain_ = false;
namespace py = pybind11;
using namespace pybind11::literals;
py::object to_float_ptr(uintptr_t input_ptr, int size, ggml_type type) {
if (type < 0 || type >= GGML_TYPE_COUNT) {
PyErr_SetString(PyExc_ValueError, "Invalid ggml_type");
@ -473,7 +474,6 @@ void bind_moe_module(py::module_& moe_module, const char* name) {
}
PYBIND11_MODULE(kt_kernel_ext, m) {
py::class_<WorkerPool>(m, "WorkerPool").def(py::init<int>());
py::class_<WorkerPoolConfig>(m, "WorkerPoolConfig")
.def(py::init<>())
@ -813,7 +813,7 @@ PYBIND11_MODULE(kt_kernel_ext, m) {
bind_moe_module<AVX2_FP8_MOE_TP<avx2::GemmKernelAVX2FP8>>(moe_module, "AVX2FP8_MOE");
bind_moe_module<AVX2_GPTQ_INT4_MOE_TP<avx2::GemmKernelAVX2GPTQInt4>>(moe_module, "AVX2GPTQInt4_MOE");
bind_moe_module<AVXVNNI256_GPTQ_INT4_MOE_TP<avxvnni::GemmKernelAVXVNNI256GPTQInt4>>(moe_module,
"AVXVNNI256GPTQInt4_MOE");
"AVXVNNI256GPTQInt4_MOE");
#endif
#if defined(USE_MOE_KERNEL)
@ -976,6 +976,7 @@ PYBIND11_MODULE(kt_kernel_ext, m) {
py::arg("size"), py::arg("type"));
}
#if defined(KTRANSFORMERS_ENABLE_CPPTRACE)
static void warmup_cpptrace() {
// 避免第一次调用触发 lazy-loading（malloc 等） :contentReference[oaicite:7]{index=7}
cpptrace::frame_ptr buffer[10];
@ -1002,3 +1003,4 @@ __attribute__((constructor)) static void install_handlers() {
sigaction(SIGABRT, &sa, nullptr);
}
#endif