This change copies the cmd/nvidia-container-runtime, internal, and test
folders from github.com/NVIDIA/nvidia-container-runtime@8a63b4b34f3ce3b4167f0516aa3f7207ca280dfb
Signed-off-by: Evan Lezar <elezar@nvidia.com>
This change adds support for the NVIDIA_FABRIC_DEVICES envvar. The (non-empty)
value of this envvar is passed to the NVIDIA Container CLI using the --fabric-device
command line flag and allows for nvswitch and nvlink devices to be mounted
into the container.
Signed-off-by: Evan Lezar <elezar@nvidia.com>
This change improves the CI for the container toolkit. The go targets are
executed in a docker container which allows for reproducible behaviour on
local systems as well as CI. The Makefile is updated to facilitate this.
Signed-off-by: Evan Lezar <elezar@nvidia.com>
This change replaces the reference to gitlab.com/nvidia/cloud-native/container-toolkit
with github.com/NVIDIA/nvidia-container-toolkit matching this project.
Signed-off-by: Evan Lezar <elezar@nvidia.com>
This change moves the pkg folder to `cmd/nvidia-container-toolkit` to
better match go best practices. This allows, for example, for the
`cmd/nvidia-container-toolkit` to be go installed.
The only package included in `pkg` was `main`.
Signed-off-by: Evan Lezar <elezar@nvidia.com>
This change fixes a bug where the value of NVIDIA_VISIBLE_DEVICES would be used to
select devices even if the `swarm-resource` config option is specified.
Note that this does not change the value of NVIDIA_VISIBLE_DEVICES in the container.
Signed-off-by: Evan Lezar <elezar@nvidia.com>
This change simplifies the build process by only targetting ubuntu20.04-amd64
and adds logic to push tagged builds to artifactory.
Signed-off-by: Evan Lezar <elezar@nvidia.com>
This change adds check targets for Golang to the make file. These are also
added as stages to the to the Jenkinsfile definition and the GitLab CI
is modified to use them too.
Signed-off-by: Evan Lezar <elezar@nvidia.com>
This change ignores the value of NVIDIA_VISIBLE_DEVICES instead of
raising an error when launching a container with insufficient permissions.
This changes the behaviour under the following conditions:
NVIDIA_VISIBLE_DEVICES is set
and
accept-nvidia-visible-devices-envvar-when-unprivileged = false (default: true)
or
privileged = false (default: false)
This means that a user need not explicitly clear the NVIDIA_VISIBLE_DEVICES
environment variable if no GPUs are to be used in unprivileged containers.
Note that this envvar is set to 'all' by default in many CUDA images that
are used as base images.
Signed-off-by: Evan Lezar <elezar@nvidia.com>
