UniqAI/nvidia-container-toolkit
1
0
Fork 0
mirror of https://github.com/NVIDIA/nvidia-container-toolkit synced 2025-06-26 18:18:24 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Add cuda-compat-mode config option

Browse Source 
This change adds an nvidia-container-runtime.modes.legacy.cuda-compat-mode
config option. This can be set to one of four values:

* ldconfig (default): the --cuda-compat-mode=ldconfig flag is passed to the nvidia-container-cli
* mount: the --cuda-compat-mode=mount flag is passed to the nvidia-conainer-cli
* disabled: the --cuda-compat-mode=disabled flag is passed to the nvidia-container-cli
* hook: the --cuda-compat-mode=disabled flag is passed to the nvidia-container-cli AND the
  enable-cuda-compat hook is used to provide forward compatibility.

Note that the disable-cuda-compat-lib-hook feature flag will prevent the enable-cuda-compat
hook from being used. This change also means that the allow-cuda-compat-libs-from-container
feature flag no longer has any effect.

Signed-off-by: Evan Lezar <elezar@nvidia.com>
This commit is contained in:
Evan Lezar
2025-04-29 14:00:12 +02:00
parent 34526b19c0
commit 8069346746
7 changed files with 118 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files

33
internal/config/runtime.go
View File

@@ -29,8 +29,9 @@ type RuntimeConfig struct {
// modesConfig defines (optional) per-mode configs
type modesConfig struct {
CSV csvModeConfig `toml:"csv"`
CDI cdiModeConfig `toml:"cdi"`
CSV csvModeConfig `toml:"csv"`
CDI cdiModeConfig `toml:"cdi"`
Legacy legacyModeConfig `toml:"legacy"`
}
type cdiModeConfig struct {
@@ -45,3 +46,31 @@ type cdiModeConfig struct {
type csvModeConfig struct {
MountSpecPath string `toml:"mount-spec-path"`
}
type legacyModeConfig struct {
// CUDACompatMode sets the mode to be used to make CUDA Forward Compat
// libraries discoverable in the container.
CUDACompatMode cudaCompatMode `toml:"cuda-compat-mode,omitempty"`
}
type cudaCompatMode string
const (
defaultCUDACompatMode = CUDACompatModeLdconfig
// CUDACompatModeDisabled explicitly disables the handling of CUDA Forward
// Compatibility in the NVIDIA Container Runtime and NVIDIA Container
// Runtime Hook.
CUDACompatModeDisabled = cudaCompatMode("disabled")
// CUDACompatModeHook uses a container lifecycle hook to implement CUDA
// Forward Compatibility support. This requires the use of the NVIDIA
// Container Runtime and is not compatible with use cases where only the
// NVIDIA Container Runtime Hook is used (e.g. the Docker --gpus flag).
CUDACompatModeHook = cudaCompatMode("hook")
// CUDACompatModeLdconfig adds the folders containing CUDA Forward Compat
// libraries to the ldconfig command invoked from the NVIDIA Container
// Runtime Hook.
CUDACompatModeLdconfig = cudaCompatMode("ldconfig")
// CUDACompatModeMount mounts CUDA Forward Compat folders from the container
// to the container when using the NVIDIA Container Runtime Hook.
CUDACompatModeMount = cudaCompatMode("mount")
)