Evan LezarandPiotr Mlocek 047de66b22 feat(bootstrap,cli): switch GPU injection to CDI where supported (#495) ...

* feat(bootstrap): switch GPU injection to CDI where supported

Use an explicit CDI device request (driver="cdi", device_ids=["nvidia.com/gpu=all"])
when the Docker daemon reports CDI spec directories via GET /info (SystemInfo.CDISpecDirs).
This makes device injection declarative and decouples spec generation from consumption.

When the daemon reports no CDI spec directories, fall back to the legacy NVIDIA device
request (driver="nvidia", count=-1) which relies on the NVIDIA Container Runtime hook.
Failure modes for both paths are equivalent: a missing or stale NVIDIA Container Toolkit
installation will cause container start to fail.

CDI spec generation is out of scope for this change; specs are expected to be
pre-generated out-of-band, for example by the NVIDIA Container Toolkit.

---------

Signed-off-by: Evan Lezar <elezar@nvidia.com>
Co-authored-by: Piotr Mlocek <pmlocek@nvidia.com>

2026-03-31 08:23:52 -07:00

..

src

fix(bootstrap): detect missing sandbox supervisor binary during gateway health check (#281)

2026-03-13 15:07:50 -07:00

tests

feat(bootstrap,cli): switch GPU injection to CDI where supported (#495)

2026-03-31 08:23:52 -07:00

Cargo.lock

refactor(python): rename navigator module to openshell and migrate config to gateway paths (#220)

2026-03-10 22:24:04 -07:00

Cargo.toml

chore: rename project from NemoClaw to OpenShell (#198)

2026-03-10 11:49:09 -07:00