NVIDIA / nvidia-container-toolkit
44%
main: 44%

LAST BUILD BRANCH: cuda-memory-limits-impl
DEFAULT BRANCH: main
Repo Added
Token yURQA9qJZkSTLsagBnXzxNupHtB5LQLk7 regen
Build 3240 Last
Files 250
Badge
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst

LAST BUILD ON BRANCH pull-request/2074
branch: pull-request/2074
CHANGE BRANCH
x
Sync Branches

coverage: 44.057% (+0.007%) from 44.05%
34539754510

Pull #2074

github

web-flow 
update-ldcache hook: exec ldconfig from a sealed memfd

UpdateLDCache used to mount a fresh /proc inside the container root,
just so it could bind-mount the host ldconfig binary in through a
TOCTOU-safe /proc/self/fd/<n> target and then exec it by that same
path once pivoted.

This breaks under gVisor. Its createContainer hook doesn't have
permission to mount a new /proc inside the container, so mountProc
fails with EPERM. This constitutes a real regression, because this
hook worked fine before the changes described above were introduced.

To address this, UpdateLDCache now clones the host ldconfig binary
into a sealed, anonymous memfd (via libcontainer/exeseal.CloneBinary)
before touching any namespace or pivoting, then execs it by
descriptor with execveat(fd, "", ..., AT_EMPTY_PATH). A memfd has no
path and no mount-namespace membership, so it comes through
pivot_root untouched. We don't need to bind-mount anything into the
container, and we don't need to mount /proc anywhere.

This is also more isolated than before. Since ldconfig has no use
for /proc or /sys, prepareRoot now masks both with an empty tmpfs
before pivoting, if they're present in the container root, instead of
leaving a real working procfs mounted there and trusting ldconfig
never reads it.

We verified all of this against real GPU hardware through gVisor's
runsc runtime, as well as standard runc. `docker run --runtime=runsc
--device nvidia.com/gpu=0` against a non-Debian image resolves the
driver libraries and runs nvidia-smi correctly, confirming the hook
and the /proc/sys masking both work under gVisor's restricted-
namespace gofer process.

Signed-off-by: Kevin Klues <kklues@nvidia.com>
(cherry picked from commit 111337006)
Pull Request #2074: [release-1.20] update-ldcache hook: exec ldconfig from a sealed memfd

0 of 58 new or added lines in 2 files covered. (0.0%)

5 existing lines in 2 files now uncovered.

5960 of 13528 relevant lines covered (44.06%)

0.44 hits per line

Relevant lines Covered
13528 RELEVANT LINES 5960 COVERED LINES
0.44 HITS PER LINE
Source Files on pull-request/2074

Recent builds

Builds Branch Commit Type Ran Committer Via Coverage
34539754510 pull-request/2074 update-ldcache hook: exec ldconfig from a sealed memfd UpdateLDCache used to mount a fresh /proc inside the container root, just so it could bind-mount the host ldconfig binary in through a TOCTOU-safe /proc/self/fd/<n> target and then exec it by... Pull #2074 web-flow github
44.06
See All Builds (3073)

Badge your Repo: nvidia-container-toolkit

We detected this repo isn’t badged! Grab the embed code to the right, add it to your repo to show off your code coverage, and when the badge is live hit the refresh button to remove this message.
Embed ▾
README BADGES
x

If you need to use a raster PNG badge, change the '.svg' to '.png' in the link

Markdown

Textile

RDoc

HTML

Rst

Refresh