NVIDIA / nvidia-container-toolkit / 35137579073

44%

main: 44%

Build Type

Pull #2104

github

Committed by web-flow

Commit Message

fix: [apply-cuda-memory-limits] exit gracefully on oci spec load failure

With this fix, containers started with "userns --nomap" no longer fail as
the OCI spec load operation is no longer non-blocking. This is required as
the containers with "userns --nomap" fail even if they don't set the cuda
memory limits environment variables.

Signed-off-by: Tariq Ibrahim <tibrahim@nvidia.com>
(cherry picked from commit 1e5fb484c)

Pull Request Pull Request #2104: [release-1.20] fix: [apply-cuda-memory-limits] exit gracefully on oci spec load failure

Coverage Stats

0 of 2 new or added lines in 1 file covered. (0.0%)

1 existing line in 1 file now uncovered.

6041 of 13693 relevant lines covered (44.12%)

0.44 hits per line