Go to file
Evan Lezar fc65d3a784
Add enable-cuda-compat hook to allow compat libs to be discovered
This change adds an nvidia-cdi-hook enable-cuda-compat hook that checks the
container for cuda compat libs and updates /etc/ld.so.conf.d to include their
parent folder if their driver major version is sufficient.

This allows CUDA Forward Compatibility to be used when this is not available
through the libnvidia-container.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
2025-02-27 15:58:15 +02:00
.github [no-relnote] Update Slack notification action step 2025-02-13 14:16:44 +01:00
cmd Add enable-cuda-compat hook to allow compat libs to be discovered 2025-02-27 15:58:15 +02:00
deployments Merge pull request #925 from NVIDIA/dependabot/docker/deployments/devel/main/golang-1.24.0 2025-02-17 17:27:25 +01:00
docker
hack
internal Allow cdi mode to work with --gpus flag 2025-02-05 19:01:43 +01:00
packaging
pkg Disable nvsandboxutils in nvcdi API 2025-02-26 14:45:22 +02:00
scripts
testdata
tests Bump golang.org/x/crypto from 0.33.0 to 0.35.0 in /tests 2025-02-25 09:12:04 +00:00
third_party Bump third_party/libnvidia-container from f23e5e5 to 95d3e86 2025-01-24 09:05:17 +00:00
vendor Bump tags.cncf.io/container-device-interface from 0.8.0 to 0.8.1 2025-02-25 09:11:59 +00:00
.common-ci.yml
.dockerignore
.gitignore
.gitlab-ci.yml
.gitmodules
.golangci.yml
.nvidia-ci.yml
CHANGELOG.md Bump version for v1.17.4 release 2025-02-10 13:24:41 +01:00
CONTRIBUTING.md
DEVELOPMENT.md
go.mod Bump tags.cncf.io/container-device-interface from 0.8.0 to 0.8.1 2025-02-25 09:11:59 +00:00
go.sum Bump tags.cncf.io/container-device-interface from 0.8.0 to 0.8.1 2025-02-25 09:11:59 +00:00
LICENSE
Makefile [no-relnote] Update vendor make target to handle submodules 2025-02-12 10:56:56 +01:00
README.md
RELEASE.md
versions.mk Bump version for v1.17.4 release 2025-02-10 13:24:27 +01:00

NVIDIA Container Toolkit

GitHub license Documentation Package repository

nvidia-container-stack

Introduction

The NVIDIA Container Toolkit allows users to build and run GPU accelerated containers. The toolkit includes a container runtime library and utilities to automatically configure containers to leverage NVIDIA GPUs.

Product documentation including an architecture overview, platform support, and installation and usage guides can be found in the documentation repository.

Getting Started

Make sure you have installed the NVIDIA driver for your Linux Distribution Note that you do not need to install the CUDA Toolkit on the host system, but the NVIDIA driver needs to be installed

For instructions on getting started with the NVIDIA Container Toolkit, refer to the installation guide.

Usage

The user guide provides information on the configuration and command line options available when running GPU containers with Docker.

Issues and Contributing

Checkout the Contributing document!