Re: [PATCH v3 1/2] rtmutex: allow specifying a subclass for nested locking

From: Peter Zijlstra
Date: Mon May 28 2018 - 03:18:13 EST


On Sun, May 27, 2018 at 10:19:36PM -0700, Joel Fernandes wrote:

> > +static inline void __rt_mutex_lock(struct rt_mutex *lock, unsigned int subclass)
> > +{
> > + might_sleep();
> > +
> > + mutex_acquire(&lock->dep_map, subclass, 0, _RET_IP_);
> > + rt_mutex_fastlock(lock, TASK_UNINTERRUPTIBLE, rt_mutex_slowlock);
> > +}
> > +
> > +#ifdef CONFIG_DEBUG_LOCK_ALLOC
> > +/**
> > + * rt_mutex_lock_nested - lock a rt_mutex
>
> This ifdef seems consistent with other nested locking primitives, but its
> kind of confusing.
>
> The Kconfig.debug for DEBUG_LOCK_ALLOC says:
>
> config DEBUG_LOCK_ALLOC
> bool "Lock debugging: detect incorrect freeing of live locks"
> [...]
> help
> This feature will check whether any held lock (spinlock, rwlock,
> mutex or rwsem) is incorrectly freed by the kernel, via any of the
> memory-freeing routines (kfree(), kmem_cache_free(), free_pages(),
> vfree(), etc.), whether a live lock is incorrectly reinitialized via
> spin_lock_init()/mutex_init()/etc., or whether there is any lock
> held during task exit.
>
> Shouldn't this ideally be ifdef'd under PROVE_LOCKING for this and other
> locking primitives? Any idea what's the reason? I know PROVE_LOCKING selects
> DEBUG_LOCK_ALLOC but still..

No, the reason is that DEBUG_LOCK_ALLOC needs the lockdep hooks to know
which locks are held, so it can warn when we try and free a held one.
PROVE_LOCKING builds upon that.

The the locking primitives should key off of DEBUG_LOCK_ALLOC for
introducing the hooks.