This change adds CDI device IDs to the ContainerAllocateResponse in the
device plugin API. This allows a device plugin to specify CDI devices
by their unique fully-qualified CDI device names using the related field
in the CRI specification.
Signed-off-by: Evan Lezar <elezar@nvidia.com>
Kubernetes-commit: b57c7e2fe4bb466ff1614aa9df7cc164e90b24b6
Each of these scripts is basically identical, and all were too brittle.
Now they should be more resilient and easier to manage. The script
still needs to be updated if we add new ones, which I do not love.
More cleanup to follow.
Kubernetes-commit: e0ecccff3f5148cc167117ac73233b4edc1640d8
- Run hack/update-codegen.sh
- Run hack/update-generated-device-plugin.sh
- Run hack/update-generated-protobuf.sh
- Run hack/update-generated-runtime.sh
- Run hack/update-generated-swagger-docs.sh
- Run hack/update-openapi-spec.sh
- Run hack/update-gofmt.sh
Signed-off-by: Davanum Srinivas <davanum@gmail.com>
Kubernetes-commit: a9593d634c6a053848413e600dadbf974627515f
The protocol buffer [styleguide] says field names should use
underscore_separated_names.
Code generation tools rely on this convention to split words and
generate language-specific accessors/properties in the local style
conventions (ie: it's more than just a style convention). For
example, the previous `deviceIDs` name caused Rust proto tools to
generate field names like `device_i_ds` which is just weird.
[styleguide]: https://developers.google.com/protocol-buffers/docs/style#message-and-field-names
Rename `deviceIDs` field to `device_ids`, and use a gogo-specific
option to keep the existing `DeviceIDs` golang name (otherwise it
would generate `DeviceIds`).
Note this doesn't affect protocol buffer / gRPC wire encoding, which
uses numeric values.
Kubernetes-commit: 815dcd5b3174f9ccaa3789f1b688a809223ca860