Skip to content

docs(cuda.core): correct the 1.1.1 program cache permissions note #2718

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
mdboom merged 3 commits into from
Oct 30, 2026
Merged

docs(cuda.core): correct the 1.1.1 program cache permissions note #2718

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
586e1ba
docs(cuda.core): correct the 1.1.1 program cache permissions note
fedonman Aug 27, 2026
acaf6b5
Merge branch 'main' into docs-1-1-1-program-cache-perms
fedonman Sep 26, 2026
fd0f548
Merge branch 'main' into docs-1-1-1-program-cache-perms
mdboom Oct 30, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 6 additions & 4 deletions cuda_core/docs/source/release/1.1.1-notes.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,10 +41,12 @@ Fixes and enhancements
internal driver thread.
(`#2371 <https://github.com/NVIDIA/cuda-python/pull/2371>`__)

- The on-disk program cache directory is now created with owner-only
permissions (``0o700``) on POSIX systems, and those permissions are
re-asserted on each use. This prevents other local users from reading or
injecting cached device code regardless of the process ``umask``.
- Program cache entry files are now written with owner-only permissions
(``0o600``) on POSIX systems, and the ``tmp/`` staging directory is created
``0o700``, so other local users cannot read cached device code regardless of
the process ``umask``. The cache root and ``entries/`` inherit the umask and
a pre-existing cache directory is used as-is, so a deliberately shared cache
Comment on lines +46 to +48

@coderabbitai coderabbitai Bot Oct 30, 2026

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Minor · Security & Privacy · Quick win

important: Scope the confidentiality claim to newly written files.

FileStreamProgramCache does not chmod existing entry files or an existing tmp/ directory. A pre-existing cache can therefore contain files with broader permissions. The statement that other local users cannot read cached device code is too broad. State that the 0o600 and 0o700 guarantees apply when the files and directories are created, while existing cache contents remain unchanged.

keeps working.
(`#2399 <https://github.com/NVIDIA/cuda-python/pull/2399>`__)

- DLPack: a ``NULL`` deleter in a ``DLManagedTensorVersioned`` capsule is now
Expand Down
Loading