| Age | Commit message (Collapse) | Author |
|---|
|
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This is not possible to use directly, so remove it from the public API.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This is subtle to use correctly. It also does not make sense as part of the container API.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This turned out not to require specific treatment, so the shared interfaces are cleaner.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
The go116 behaviour of built-in new function makes this cleaner.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This prevents inadvertent use of insecure compatibility features.
Closes #30.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This change also renames confusingly named flags field and corrects its doc comment.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This also removes arch-specific patches because they were not useful.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This now passes the test suite.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
Some aarch64 regression tests fail intermittently on the default size.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This deals with environment variables generally present in every message.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This works around the race in vfs via supplementary group.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This happens in the vfs permissions check only and stale data appears to never reach userspace.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This works around kernels with Landlock LSM disabled. Does not affect cure outcome.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This is quite expensive, but no other option, unfortunately.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This usually ends up in the buffer, or completes well before the deadline, however this can still timeout on a very slow system.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This does not make sense as part of the public API, so make it internal.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
The API forces use of finalizer to close the read end of the setup pipe, which is no longer considered acceptable. Exporting this as part of package container also imposes unnecessary maintenance burden.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This enables transparently degradable tests to be forced on in environments known to support them.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
Explicitly requiring landlock in tests will be supported in a future change.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
The program generally never blocks, and it is more flexible to leave it up to the caller to set a timeout.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This does not test package container, so should transparently cope with Landlock LSM being unavailable.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This overlaps with net namespace, so can be skipped without degrading security.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
The alpine linux riscv64 kernel does not enable Landlock LSM, and kernel compilation is not yet feasible.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This has no effect, but is nice to have.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This had not yet passed all test suites because emulator is prohibitively slow.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This enables rejection of non-coldboot synthetic events.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This is much more robust than a shell script.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This breaks on shells that do not evaluate pathnames.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This enables the writable root overlay use case.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This no longer relies on the hermetic build system to clean up.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
These are baked in, so make them configurable for the build.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This is not considered an error, if the process finishes while the signal is being delivered.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
Eliminates the xz dependency.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
Eliminates the xz dependency.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
Eliminates the xz dependency.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
Eliminates the xz dependency.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
Eliminates the xz dependency.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
Eliminates the xz dependency.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
Eliminates the xz dependency. This also switches to meson to avoid pulling in autotools.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
This eliminates xz dependency.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
Hardware on arm64 is quite messy, this miscompiles.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
The test suite is somehow happy on amd64 but fails on arm64.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
|
The .git directory is destroyed anyway, so no point fetching more than the bare minimum.
Signed-off-by: Ophestra <cat@gensokyo.uk>
|
