docker.io/javirub/flashattention-pytorch:flashattn2.8.1-pytorch2.7.1-cuda12.8-cudnn9-runtime linux/amd64

Docker拉取命令

docker pull swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/javirub/flashattention-pytorch:flashattn2.8.1-pytorch2.7.1-cuda12.8-cudnn9-runtime
docker tag  swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/javirub/flashattention-pytorch:flashattn2.8.1-pytorch2.7.1-cuda12.8-cudnn9-runtime  docker.io/javirub/flashattention-pytorch:flashattn2.8.1-pytorch2.7.1-cuda12.8-cudnn9-runtime

Containerd拉取命令

ctr images pull swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/javirub/flashattention-pytorch:flashattn2.8.1-pytorch2.7.1-cuda12.8-cudnn9-runtime
ctr images tag  swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/javirub/flashattention-pytorch:flashattn2.8.1-pytorch2.7.1-cuda12.8-cudnn9-runtime  docker.io/javirub/flashattention-pytorch:flashattn2.8.1-pytorch2.7.1-cuda12.8-cudnn9-runtime

Shell快速替换命令

sed -i 's#javirub/flashattention-pytorch:flashattn2.8.1-pytorch2.7.1-cuda12.8-cudnn9-runtime#swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/javirub/flashattention-pytorch:flashattn2.8.1-pytorch2.7.1-cuda12.8-cudnn9-runtime#' deployment.yaml

Ansible快速分发-Docker

#ansible k8s -m shell -a 'docker pull swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/javirub/flashattention-pytorch:flashattn2.8.1-pytorch2.7.1-cuda12.8-cudnn9-runtime && docker tag  swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/javirub/flashattention-pytorch:flashattn2.8.1-pytorch2.7.1-cuda12.8-cudnn9-runtime  docker.io/javirub/flashattention-pytorch:flashattn2.8.1-pytorch2.7.1-cuda12.8-cudnn9-runtime'

Ansible快速分发-Containerd

#ansible k8s -m shell -a 'ctr images pull swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/javirub/flashattention-pytorch:flashattn2.8.1-pytorch2.7.1-cuda12.8-cudnn9-runtime && ctr images tag  swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/javirub/flashattention-pytorch:flashattn2.8.1-pytorch2.7.1-cuda12.8-cudnn9-runtime  docker.io/javirub/flashattention-pytorch:flashattn2.8.1-pytorch2.7.1-cuda12.8-cudnn9-runtime'

镜像构建历史

# 2025-07-15 20:07:00  0.00B 设置默认要执行的命令
CMD ["bash"]
                        
# 2025-07-15 20:07:00  0.00B 执行命令并创建新的镜像层
RUN /bin/sh -c python -c "import flash_attn; print(f'FlashAttention version: {flash_attn.__version__}')" # buildkit
                        
# 2025-07-15 20:06:58  140.74MB 执行命令并创建新的镜像层
RUN /bin/sh -c pip install /tmp/wheels/flash_attn-*.whl &&     rm -rf /tmp/wheels &&     rm -rf /root/.cache/pip # buildkit
                        
# 2025-07-15 20:06:57  126.12MB 复制新文件或目录到容器中
COPY /wheels/flash_attn-*.whl /tmp/wheels/ # buildkit
                        
# 2025-07-15 18:04:37  37.72KB 执行命令并创建新的镜像层
RUN /bin/sh -c pip install --upgrade pip # buildkit
                        
# 2025-06-05 02:24:02  0.00B 设置工作目录为/workspace
WORKDIR /workspace
                        
# 2025-06-05 02:24:02  0.00B 设置环境变量 PYTORCH_VERSION
ENV PYTORCH_VERSION=2.7.1
                        
# 2025-06-05 02:24:02  0.00B 设置环境变量 PATH
ENV PATH=/usr/local/nvidia/bin:/usr/local/cuda/bin:/opt/conda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
                        
# 2025-06-05 02:24:02  0.00B 设置环境变量 LD_LIBRARY_PATH
ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64
                        
# 2025-06-05 02:24:02  0.00B 设置环境变量 NVIDIA_DRIVER_CAPABILITIES
ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility
                        
# 2025-06-05 02:24:02  0.00B 设置环境变量 NVIDIA_VISIBLE_DEVICES
ENV NVIDIA_VISIBLE_DEVICES=all
                        
# 2025-06-05 02:24:02  0.00B 设置环境变量 PATH
ENV PATH=/opt/conda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
                        
# 2025-06-05 02:24:02  0.00B 执行命令并创建新的镜像层
RUN |4 PYTORCH_VERSION=2.7.1 TRITON_VERSION= TARGETPLATFORM=linux/amd64 CUDA_VERSION=12.8.0 /bin/sh -c if test -n "${TRITON_VERSION}" -a "${TARGETPLATFORM}" != "linux/arm64"; then         DEBIAN_FRONTEND=noninteractive apt install -y --no-install-recommends gcc;         rm -rf /var/lib/apt/lists/*;     fi # buildkit
                        
# 2025-06-05 02:24:01  7.49GB 复制新文件或目录到容器中
COPY /opt/conda /opt/conda # buildkit
                        
# 2025-06-05 02:19:30  25.98MB 执行命令并创建新的镜像层
RUN |4 PYTORCH_VERSION=2.7.1 TRITON_VERSION= TARGETPLATFORM=linux/amd64 CUDA_VERSION=12.8.0 /bin/sh -c apt-get update && DEBIAN_FRONTEND=noninteractive apt-get install -y --no-install-recommends         ca-certificates         libjpeg-dev         libpng-dev         && rm -rf /var/lib/apt/lists/* # buildkit
                        
# 2025-06-05 02:19:30  0.00B 添加元数据标签
LABEL com.nvidia.volumes.needed=nvidia_driver
                        
# 2025-06-05 02:19:30  0.00B 定义构建参数
ARG CUDA_VERSION=12.8.0
                        
# 2025-06-05 02:19:30  0.00B 定义构建参数
ARG TARGETPLATFORM=linux/amd64
                        
# 2025-06-05 02:19:30  0.00B 定义构建参数
ARG TRITON_VERSION=
                        
# 2025-06-05 02:19:30  0.00B 定义构建参数
ARG PYTORCH_VERSION=2.7.1
                        
# 2025-05-31 06:30:45  0.00B 
/bin/sh -c #(nop)  CMD ["/bin/bash"]
                        
# 2025-05-31 06:30:45  77.86MB 
/bin/sh -c #(nop) ADD file:82f38ebced7b2756311fb492d3d44cc131b22654e8620baa93883537a3e355aa in / 
                        
# 2025-05-31 06:30:42  0.00B 
/bin/sh -c #(nop)  LABEL org.opencontainers.image.version=22.04
                        
# 2025-05-31 06:30:42  0.00B 
/bin/sh -c #(nop)  LABEL org.opencontainers.image.ref.name=ubuntu
                        
# 2025-05-31 06:30:42  0.00B 
/bin/sh -c #(nop)  ARG LAUNCHPAD_BUILD_ARCH
                        
# 2025-05-31 06:30:42  0.00B 
/bin/sh -c #(nop)  ARG RELEASE
                        
                    

镜像信息

{
    "Id": "sha256:73009848e7433eea76527ac97e12a4264c1e20cf8ba6e1edb6477c7889306cd4",
    "RepoTags": [
        "javirub/flashattention-pytorch:flashattn2.8.1-pytorch2.7.1-cuda12.8-cudnn9-runtime",
        "swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/javirub/flashattention-pytorch:flashattn2.8.1-pytorch2.7.1-cuda12.8-cudnn9-runtime"
    ],
    "RepoDigests": [
        "javirub/flashattention-pytorch@sha256:0db4d8db06003e1983da398a8923a48283c13fce0e9fbf167d7d7e554d970952",
        "swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/javirub/flashattention-pytorch@sha256:826e63538ee9775627cdf232bf6caa997a43bd1889901664a8762032be031614"
    ],
    "Parent": "",
    "Comment": "buildkit.dockerfile.v0",
    "Created": "2025-07-15T12:07:00.784919815Z",
    "Container": "",
    "ContainerConfig": null,
    "DockerVersion": "",
    "Author": "",
    "Config": {
        "Hostname": "",
        "Domainname": "",
        "User": "",
        "AttachStdin": false,
        "AttachStdout": false,
        "AttachStderr": false,
        "Tty": false,
        "OpenStdin": false,
        "StdinOnce": false,
        "Env": [
            "PATH=/usr/local/nvidia/bin:/usr/local/cuda/bin:/opt/conda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin",
            "NVIDIA_VISIBLE_DEVICES=all",
            "NVIDIA_DRIVER_CAPABILITIES=compute,utility",
            "LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64",
            "PYTORCH_VERSION=2.7.1"
        ],
        "Cmd": [
            "bash"
        ],
        "ArgsEscaped": true,
        "Image": "",
        "Volumes": null,
        "WorkingDir": "/workspace",
        "Entrypoint": null,
        "OnBuild": null,
        "Labels": {
            "com.nvidia.volumes.needed": "nvidia_driver",
            "org.opencontainers.image.ref.name": "ubuntu",
            "org.opencontainers.image.version": "22.04"
        }
    },
    "Architecture": "amd64",
    "Os": "linux",
    "Size": 7864845060,
    "GraphDriver": {
        "Data": {
            "LowerDir": "/var/lib/docker/overlay2/e65a071dd64bf9bbeb19f60fd4765bfafa3c4f7e0defa77fde2597d2e58415fe/diff:/var/lib/docker/overlay2/205cd079b004263491f23ffe08cd2f27a0338e25158d1b16089ea127de217ec0/diff:/var/lib/docker/overlay2/debbbf43a9344b1c5085efb5d8c2f119cc61974736af03f467c4badbd69e66a4/diff:/var/lib/docker/overlay2/0d0408ee2a819f69781460b9a1f162a4b70e743a1258f060c5f11ed8a49eba16/diff:/var/lib/docker/overlay2/773a7e689c208300aa4a7a369635ca7761c232255a661d9e4e55ec130915d310/diff:/var/lib/docker/overlay2/e65a0581a17bb3c604d8973577fff4018d81044514237920371d2bbd24603971/diff:/var/lib/docker/overlay2/299d50cce2b2ed4927736f0c8044f6cd6fcfcaf8f250a49f5ae9f99ec1c2af45/diff:/var/lib/docker/overlay2/e8ecf5439449dc7758d2a4396a5d5b5efd20bf2f377e87c97b3a2db49476c293/diff",
            "MergedDir": "/var/lib/docker/overlay2/96f98170edc2dd52c3b1589e211e632d1786f7d1297798f234145ff056de9299/merged",
            "UpperDir": "/var/lib/docker/overlay2/96f98170edc2dd52c3b1589e211e632d1786f7d1297798f234145ff056de9299/diff",
            "WorkDir": "/var/lib/docker/overlay2/96f98170edc2dd52c3b1589e211e632d1786f7d1297798f234145ff056de9299/work"
        },
        "Name": "overlay2"
    },
    "RootFS": {
        "Type": "layers",
        "Layers": [
            "sha256:f862e1968e4b4c3c3af141e37d2ec22b19ec0fd50d6a8aaf683de6729e296226",
            "sha256:105c4058ec6f14215f0da98eaeae0dac3960e697ee19b38f2d8a365e679ec974",
            "sha256:2bf9ca7c9c371a3db2515b77c08c0f85c5c03821a2c383861680ee6dad1b427e",
            "sha256:5f70bf18a086007016e948b04aed3b82103a36bea41755b6cddfaf10ace3c6ef",
            "sha256:1fda46049be10a2677981d12a76701bd60a254a61414f0e7e97fa1e0cf45064f",
            "sha256:373b38029eadde31283d8f4bb6f89f2963f2136f8c18096dee5085f5e80d8824",
            "sha256:36be649a84a7779b1f3815256dae0ee8a5fe10d8199ba88534b2e41ddd9dafb9",
            "sha256:64422af63e49f8e9306afd50115e301aa4a5f40f6d5a3f988b8ab29f839da1b3",
            "sha256:5f70bf18a086007016e948b04aed3b82103a36bea41755b6cddfaf10ace3c6ef"
        ]
    },
    "Metadata": {
        "LastTagTime": "2026-06-10T00:23:58.618394276+08:00"
    }
}