CVE-2025-40231

In the Linux kernel, the following vulnerability has been resolved:

vsock: fix lock inversion in vsockassigntransport()

Syzbot reported a potential lock inversion deadlock between vsockregistermutex and sklock-AFVSOCK when vsock_linger() is called.

The issue was introduced by commit 687aa0c5581b ("vsock: Fix transport_* TOCTOU") which added vsockregistermutex locking in vsockassigntransport() around the transport->release() call, that can call vsocklinger(). vsockassigntransport() can be called with sklock held. vsocklinger() calls skwaitevent() that temporarily releases and re-acquires sklock. During this window, if another thread hold vsockregistermutex while trying to acquire sk_lock, a circular dependency is created.

Fix this by releasing vsockregistermutex before calling transport->release() and vsockdeassigntransport(). This is safe because we don't need to hold vsockregistermutex while releasing the old transport, and we ensure the new transport won't disappear by obtaining a module reference first via trymoduleget().