-
Notifications
You must be signed in to change notification settings - Fork 32
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Signed-off-by: shiva kumar <[email protected]>
- Loading branch information
Showing
12 changed files
with
1,013 additions
and
1 deletion.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,111 @@ | ||
FROM nvcr.io/nvidia/cuda:12.6.2-base-amzn2023 AS build | ||
|
||
ARG TARGETARCH | ||
|
||
SHELL ["/bin/bash", "-c"] | ||
|
||
# Remove cuda repository to avoid GPG errors | ||
RUN rm -f /etc/yum.repos.d/cuda* | ||
|
||
RUN dnf update -y && dnf makecache && \ | ||
dnf install -y \ | ||
gcc \ | ||
gcc-c++ \ | ||
make \ | ||
ca-certificates \ | ||
git \ | ||
tar && \ | ||
dnf clean all && rm -rf /var/cache/yum/* | ||
|
||
ENV GOLANG_VERSION=1.23.2 | ||
|
||
# download appropriate binary based on the target architecture for multi-arch builds | ||
RUN curl https://storage.googleapis.com/golang/go${GOLANG_VERSION}.linux-${TARGETARCH}.tar.gz \ | ||
| tar -C /usr/local -xz | ||
|
||
ENV PATH /usr/local/bin:$PATH | ||
ENV PATH /usr/local/go/bin:$PATH | ||
|
||
WORKDIR /work | ||
|
||
RUN git clone https://github.com/NVIDIA/gpu-driver-container driver && \ | ||
cd driver/vgpu/src && \ | ||
go build -o vgpu-util && \ | ||
mv vgpu-util /work | ||
|
||
FROM nvcr.io/nvidia/cuda:12.6.2-base-amzn2023 | ||
|
||
SHELL ["/bin/bash", "-c"] | ||
|
||
ARG BASE_URL=https://us.download.nvidia.com/tesla | ||
ARG TARGETARCH | ||
ENV TARGETARCH=$TARGETARCH | ||
ENV DRIVER_ARCH=${TARGETARCH/amd64/x86_64} | ||
ARG DRIVER_VERSION | ||
ENV DRIVER_VERSION=$DRIVER_VERSION | ||
|
||
# Arg to indicate if driver type is either of passthrough(baremetal) or vgpu | ||
ARG DRIVER_TYPE=passthrough | ||
ENV DRIVER_TYPE=$DRIVER_TYPE | ||
ARG DRIVER_BRANCH=560 | ||
ENV DRIVER_BRANCH=$DRIVER_BRANCH | ||
ARG VGPU_LICENSE_SERVER_TYPE=NLS | ||
ENV VGPU_LICENSE_SERVER_TYPE=$VGPU_LICENSE_SERVER_TYPE | ||
# Enable vGPU version compability check by default | ||
ARG DISABLE_VGPU_VERSION_CHECK=true | ||
ENV DISABLE_VGPU_VERSION_CHECK=$DISABLE_VGPU_VERSION_CHECK | ||
ENV NVIDIA_VISIBLE_DEVICES=void | ||
|
||
RUN echo "TARGETARCH=$TARGETARCH" | ||
|
||
ADD install.sh /tmp | ||
|
||
RUN NVIDIA_GPGKEY_SUM=d0664fbbdb8c32356d45de36c5984617217b2d0bef41b93ccecd326ba3b80c87 && \ | ||
curl -fsSL https://developer.download.nvidia.com/compute/cuda/repos/amzn2023/${DRIVER_ARCH}/D42D0685.pub | sed '/^Version/d' > /etc/pki/rpm-gpg/RPM-GPG-KEY-NVIDIA && \ | ||
echo "$NVIDIA_GPGKEY_SUM /etc/pki/rpm-gpg/RPM-GPG-KEY-NVIDIA" | sha256sum -c --strict - && \ | ||
curl -fsSL -o /etc/yum.repos.d/cuda.repo https://developer.download.nvidia.com/compute/cuda/repos/amzn2023/${DRIVER_ARCH}/cuda-amzn2023.repo | ||
|
||
RUN dnf clean all && dnf makecache && dnf update -y && dnf install -y shadow-utils | ||
RUN /tmp/install.sh reposetup && /tmp/install.sh depinstall && \ | ||
curl -fsSL -o /usr/local/bin/donkey https://github.com/3XX0/donkey/releases/download/v1.1.0/donkey && \ | ||
chmod +x /usr/local/bin/donkey | ||
|
||
COPY nvidia-driver /usr/local/bin | ||
COPY --from=build /work/vgpu-util /usr/local/bin | ||
|
||
RUN curl -fsSL -o /usr/local/bin/extract-vmlinux https://raw.githubusercontent.com/torvalds/linux/master/scripts/extract-vmlinux && \ | ||
chmod +x /usr/local/bin/extract-vmlinux | ||
|
||
ADD drivers drivers/ | ||
|
||
# Fetch the installer automatically for passthrough/baremetal types | ||
RUN if [ "$DRIVER_TYPE" != "vgpu" ]; then \ | ||
cd drivers && \ | ||
/tmp/install.sh download_installer; fi | ||
|
||
# Check for nvidia-fabric-manager or nvidia-fabricmanager availability and install | ||
RUN if [ "$DRIVER_TYPE" != "vgpu" ] && [ "$TARGETARCH" != "arm64" ]; then \ | ||
# Initialize the fabric manager package variable | ||
FABRIC_PACKAGE=""; \ | ||
if dnf list nvidia-fabric-manager-${DRIVER_VERSION}-1 &>/dev/null; then \ | ||
FABRIC_PACKAGE="nvidia-fabric-manager-${DRIVER_VERSION}-1"; \ | ||
elif dnf list nvidia-fabricmanager-${DRIVER_BRANCH}-${DRIVER_VERSION}-1 &>/dev/null; then \ | ||
FABRIC_PACKAGE="nvidia-fabricmanager-${DRIVER_BRANCH}-${DRIVER_VERSION}-1"; \ | ||
else \ | ||
echo "Error: No suitable package found for fabric manager version ${DRIVER_VERSION}"; \ | ||
exit 1; \ | ||
fi; \ | ||
dnf install -y "$FABRIC_PACKAGE" libnvidia-nscq-${DRIVER_BRANCH}-${DRIVER_VERSION}-1; fi | ||
|
||
WORKDIR /drivers | ||
|
||
ARG PUBLIC_KEY=empty | ||
COPY ${PUBLIC_KEY} kernel/pubkey.x509 | ||
|
||
# Remove cuda repository to avoid GPG errors | ||
# clean cache | ||
# Add NGC DL license from the CUDA image | ||
RUN rm -f /etc/yum.repos.d/cuda* && dnf clean all && \ | ||
mkdir /licenses && mv /NGC-DL-CONTAINER-LICENSE /licenses/NGC-DL-CONTAINER-LICENSE | ||
|
||
ENTRYPOINT ["nvidia-driver", "init"] |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,3 @@ | ||
# AmazonLinux2 [![build status](https://gitlab.com/nvidia/driver/badges/master/build.svg)](https://gitlab.com/nvidia/driver/commits/master) | ||
|
||
See https://github.com/NVIDIA/nvidia-docker/wiki/Driver-containers-(Beta) |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1 @@ | ||
# Folder for downloading vGPU drivers and dependent metadata files |
Empty file.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,58 @@ | ||
#!/usr/bin/env bash | ||
|
||
set -eu | ||
|
||
download_installer () { | ||
DRIVER_ARCH=${TARGETARCH/amd64/x86_64} && curl -fSsl -O $BASE_URL/$DRIVER_VERSION/NVIDIA-Linux-$DRIVER_ARCH-$DRIVER_VERSION.run && \ | ||
chmod +x NVIDIA-Linux-$DRIVER_ARCH-$DRIVER_VERSION.run; | ||
} | ||
|
||
dep_install () { | ||
if [ "$TARGETARCH" = "amd64" ]; then | ||
DRIVER_ARCH=${TARGETARCH/amd64/x86_64} | ||
dnf update -y && dnf install -y \ | ||
gcc \ | ||
make \ | ||
glibc-devel \ | ||
ca-certificates \ | ||
kmod \ | ||
file \ | ||
elfutils-libelf-devel \ | ||
libglvnd-devel \ | ||
shadow-utils \ | ||
util-linux \ | ||
tar \ | ||
rpm-build \ | ||
dnf-utils \ | ||
pkgconfig && \ | ||
dnf clean all && \ | ||
rm -rf /var/cache/yum/* | ||
fi | ||
} | ||
|
||
repo_setup () { | ||
if [ "$TARGETARCH" = "amd64" ]; then | ||
echo "[cuda-amzn2023-x86_64] | ||
name=cuda-amzn2023-x86_64 | ||
baseurl=https://developer.download.nvidia.com/compute/cuda/repos/amzn2023/$DRIVER_ARCH | ||
enabled=1 | ||
gpgcheck=1 | ||
gpgkey=https://developer.download.nvidia.com/compute/cuda/repos/amzn2023/$DRIVER_ARCH/D42D0685.pub" > /etc/yum.repos.d/cuda.repo && \ | ||
usermod -o -u 0 -g 0 nobody | ||
else | ||
echo "TARGETARCH doesn't match a known arch target" | ||
exit 1 | ||
fi | ||
} | ||
|
||
if [ "$1" = "reposetup" ]; then | ||
repo_setup | ||
elif [ "$1" = "depinstall" ]; then | ||
dep_install | ||
elif [ "$1" = "download_installer" ]; then | ||
download_installer | ||
else | ||
echo "Unknown function: $1" | ||
exit 1 | ||
fi | ||
|
Oops, something went wrong.