mirror of
https://github.com/kata-containers/kata-containers.git
synced 2026-07-12 01:44:30 +00:00
8f4db8a23f
Carve the monolithic NVIDIA GPU rootfs into a driver-agnostic nvidia base image (NVRC init + agent + base libs + in-tree modules) and a driver-versioned gpu extension (GPU userspace, configs, firmware, NVIDIA modules) laid out for /run/kata-extensions/gpu. Both halves are produced from the same chiseled tree via a partition-after step, so the monolith build stays byte-identical and continues to ship. The shared stage-one (driver install) is reused across the monolith, base and extension variants; nvidia base is cached without the driver/ctk versions so one base image can back multiple driver extensions. Signed-off-by: Fabiano Fidêncio <ffidencio@nvidia.com> Assisted-by: Cursor <cursoragent@cursor.com>
Design
Kata Containers design documents:
- Kata Containers architecture
- API Design of Kata Containers
- Design requirements for Kata Containers
- VSocks
- VCPU handling(in runtime-go)
- VCPU handling(in runtime-rs)
- VCPU threads pinning
- Host cgroups
- Agent systemd cgroup
InotifysupportHookssupport- Metrics(Kata 2.0)
- Metrics in Rust Runtime(runtime-rs)
- Design for Kata Containers
Lazyloadability withnydus - Design for direct-assigned volume
- Design for core-scheduling
- Virtualization Reference Architecture