LANDLOCK_RESTRICT_SELF_TSYNC does not allow
LANDLOCK_RESTRICT_SELF_LOG_SUBDOMAINS_OFF with ruleset_fd=-1, preventing
a multithreaded process from atomically propagating subdomain log muting
to all threads without creating a domain layer.  Relax the fd=-1
condition to accept TSYNC alongside LOG_SUBDOMAINS_OFF, and update the
documentation accordingly.

Add flag validation tests for all TSYNC combinations with ruleset_fd=-1,
and audit tests verifying both transition directions: muting via TSYNC
(logged to not logged) and override via TSYNC (not logged to logged).

Cc: Günther Noack <gnoack@google.com>
Cc: stable@vger.kernel.org
Fixes: 42fc7e6543f6 ("landlock: Multithreading support for landlock_restrict_self()")
Reviewed-by: Günther Noack <gnoack3000@gmail.com>
Link: https://lore.kernel.org/r/20260407164107.2012589-2-mic@digikod.net
Signed-off-by: Mickaël Salaün <mic@digikod.net>

Add tsync_interrupt test to exercise the signal interruption path in
landlock_restrict_sibling_threads().  When a signal interrupts
wait_for_completion_interruptible() while the calling thread waits for
sibling threads to finish credential preparation, the kernel:

1. Sets ERESTARTNOINTR to request a transparent syscall restart.
2. Calls cancel_tsync_works() to opportunistically dequeue task works
   that have not started running yet.
3. Breaks out of the preparation loop, then unblocks remaining
   task works via complete_all() and waits for them to finish.
4. Returns the error, causing abort_creds() in the syscall handler.

Specifically, cancel_tsync_works() in its entirety, the ERESTARTNOINTR
error branch in landlock_restrict_sibling_threads(), and the
abort_creds() error branch in the landlock_restrict_self() syscall
handler are timing-dependent and not exercised by the existing tsync
tests, making code coverage measurements non-deterministic.

The test spawns a signaler thread that rapidly sends SIGUSR1 to the
calling thread while it performs landlock_restrict_self() with
LANDLOCK_RESTRICT_SELF_TSYNC.  Since ERESTARTNOINTR causes a
transparent restart, userspace always sees the syscall succeed.

This is a best-effort coverage test: the interruption path is exercised
when the signal lands during the preparation wait, which depends on
thread scheduling.  The test creates enough idle sibling threads (200)
to ensure multiple serialized waves of credential preparation even on
machines with many cores (e.g., 64), widening the window for the
signaler.  Deterministic coverage would require wrapping the wait call
with ALLOW_ERROR_INJECTION() and using CONFIG_FAIL_FUNCTION.

Test coverage for security/landlock was 90.2% of 2105 lines according to
LLVM 21, and it is now 91.1% of 2105 lines with this new test.

Cc: Günther Noack <gnoack@google.com>
Cc: Justin Suess <utilityemal77@gmail.com>
Cc: Tingmao Wang <m@maowtm.org>
Cc: Yihan Ding <dingyihan@uniontech.com>
Link: https://lore.kernel.org/r/20260310190416.1913908-1-mic@digikod.net
Signed-off-by: Mickaël Salaün <mic@digikod.net>

Exercise various scenarios where Landlock domains are enforced across
all of a processes' threads.

Test coverage for security/landlock is 91.6% of 2130 lines according to
LLVM 21.

Cc: Andrew G. Morgan <morgan@kernel.org>
Cc: John Johansen <john.johansen@canonical.com>
Cc: Paul Moore <paul@paul-moore.com>
Signed-off-by: Günther Noack <gnoack@google.com>
Link: https://lore.kernel.org/r/20251127115136.3064948-3-gnoack@google.com
[mic: Fix subject, use EXPECT_EQ(close()), make helpers static, add test
coverage]
Signed-off-by: Mickaël Salaün <mic@digikod.net>