minke.yu/sglang
1
0
Fork 0
Code Issues Pull Requests Actions 12 Packages Projects Releases Wiki Activity

[NPU] Adapt cann 8.5: use sfa and lightning indexer op from cann and CI update (#17615)

Browse Source 
Co-authored-by: Kelon <kelonlu@163.com>
This commit is contained in:
monkeyLoveding
2026-01-27 19:03:53 +08:00
committed by GitHub
co-authored by Kelon
parent 832c756549
commit d578b41bad
9 changed files with 52 additions and 84 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docker/npu.Dockerfile
+13 -23
View File
@@ -1,4 +1,4 @@
ARG CANN_VERSION=8.3.rc2
ARG CANN_VERSION=8.5.0
ARG DEVICE_TYPE=a3
ARG OS=ubuntu22.04
ARG PYTHON_VERSION=py3.11
@@ -6,14 +6,13 @@ ARG PYTHON_VERSION=py3.11
FROM quay.io/ascend/cann:$CANN_VERSION-$DEVICE_TYPE-$OS-$PYTHON_VERSION
# Update pip & apt sources
ARG CANN_VERSION
ARG DEVICE_TYPE
ARG PIP_INDEX_URL="https://pypi.org/simple/"
ARG APTMIRROR=""
ARG PYTORCH_VERSION="2.8.0"
ARG TORCHVISION_VERSION="0.23.0"
ARG PTA_URL="https://sglang-ascend.obs.cn-east-3.myhuaweicloud.com/sglang/torch_npu/torch_npu-2.8.0.post2.dev20251113-cp311-cp311-manylinux_2_28_aarch64.whl"
ARG TRITON_ASCEND_URL="https://sglang-ascend.obs.cn-east-3.myhuaweicloud.com/sglang/triton_ascend/triton_ascend-3.2.0.dev2025112116-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl"
ARG BISHENG_NAME="Ascend-BiSheng-toolkit_aarch64_20251121.run"
ARG BISHENG_URL="https://sglang-ascend.obs.cn-east-3.myhuaweicloud.com/sglang/triton_ascend/${BISHENG_NAME}"
ARG PTA_URL="https://gitcode.com/Ascend/pytorch/releases/download/v7.3.0-pytorch2.8.0/torch_npu-2.8.0.post2-cp311-cp311-manylinux_2_28_aarch64.whl"
ARG SGLANG_TAG=main
ARG ASCEND_CANN_PATH=/usr/local/Ascend/ascend-toolkit
ARG SGLANG_KERNEL_NPU_TAG=main
@@ -31,6 +30,7 @@ RUN if [ -n "$APTMIRROR" ];then sed -i "s|.*.ubuntu.com|$APTMIRROR|g" /etc/apt/s
# Install development tools and utilities
RUN apt-get update -y && apt upgrade -y && apt-get install -y \
unzip \
build-essential \
cmake \
vim \
@@ -57,7 +57,7 @@ ENV LC_ALL=en_US.UTF-8
### Install MemFabric
RUN ${PIP_INSTALL} memfabric-hybrid==1.0.0
RUN ${PIP_INSTALL} memfabric-hybrid==1.0.5
### Install SGLang Model Gateway
RUN ${PIP_INSTALL} sglang-router
@@ -69,7 +69,7 @@ RUN (${PIP_INSTALL} torch==${PYTORCH_VERSION} torchvision==${TORCHVISION_VERSION
# TODO: install from pypi released triton-ascend
RUN (${PIP_INSTALL} pybind11) \
&& (${PIP_INSTALL} ${TRITON_ASCEND_URL})
&& (${PIP_INSTALL} triton-ascend)
# Install SGLang
RUN git clone https://github.com/sgl-project/sglang --branch $SGLANG_TAG && \
@@ -78,23 +78,13 @@ RUN git clone https://github.com/sgl-project/sglang --branch $SGLANG_TAG && \
# Install Deep-ep
# pin wheel to 0.45.1 ref: https://github.com/pypa/wheel/issues/662
RUN ${PIP_INSTALL} wheel==0.45.1 && git clone --branch $SGLANG_KERNEL_NPU_TAG https://github.com/sgl-project/sgl-kernel-npu.git \
&& export LD_LIBRARY_PATH=${ASCEND_CANN_PATH}/latest/runtime/lib64/stub:$LD_LIBRARY_PATH && \
source ${ASCEND_CANN_PATH}/set_env.sh && \
cd sgl-kernel-npu && \
bash build.sh \
RUN ${PIP_INSTALL} wheel==0.45.1 pybind11 pyyaml decorator scipy attrs psutil \
&& mkdir sgl-kernel-npu \
&& cd sgl-kernel-npu \
&& wget https://github.com/sgl-project/sgl-kernel-npu/releases/download/${SGLANG_KERNEL_NPU_TAG}/sgl-kernel-npu_${SGLANG_KERNEL_NPU_TAG}_${CANN_VERSION}_${DEVICE_TYPE}.zip \
&& unzip sgl-kernel-npu_${SGLANG_KERNEL_NPU_TAG}_${CANN_VERSION}_${DEVICE_TYPE}.zip \
&& ${PIP_INSTALL} output/deep_ep*.whl output/sgl_kernel_npu*.whl \
&& cd .. && rm -rf sgl-kernel-npu \
&& cd "$(python3 -m pip show deep-ep | awk '/^Location:/ {print $2}')" && ln -s deep_ep/deep_ep_cpp*.so
# Install CustomOps
RUN wget https://sglang-ascend.obs.cn-east-3.myhuaweicloud.com/ops/CANN-custom_ops-8.2.0.0-$DEVICE_TYPE-linux.aarch64.run && \
chmod a+x ./CANN-custom_ops-8.2.0.0-$DEVICE_TYPE-linux.aarch64.run && \
./CANN-custom_ops-8.2.0.0-$DEVICE_TYPE-linux.aarch64.run --quiet --install-path=/usr/local/Ascend/ascend-toolkit/latest/opp && \
wget https://sglang-ascend.obs.cn-east-3.myhuaweicloud.com/ops/custom_ops-1.0.$DEVICE_TYPE-cp311-cp311-linux_aarch64.whl && \
${PIP_INSTALL} ./custom_ops-1.0.$DEVICE_TYPE-cp311-cp311-linux_aarch64.whl
# Install Bisheng
RUN wget -O "${BISHENG_NAME}" "${BISHENG_URL}" && chmod a+x "${BISHENG_NAME}" && "./${BISHENG_NAME}" --install && rm "${BISHENG_NAME}"
&& cd "$(python3 -m pip show deep-ep | awk '/^Location:/ {print $2}')" && ln -sf deep_ep/deep_ep_cpp*.so
CMD ["/bin/bash"]