Re: [RFC][PATCH 0/4] Redesign cpu_hotplug locking.

[Ingo Molnar <mingo@elte.hu>]

* Gautham R Shenoy <ego@in.ibm.com> wrote:

> On Thu, Aug 24, 2006 at 12:34:12PM +0200, Ingo Molnar wrote:
> > Please add the appropriate lock_acquire()/lock_release() calls into the 
> > new sleeping semaphore type. Just use the parameters that rwlocks use:
> > 
> > #define rwlock_acquire(l, s, t, i)            lock_acquire(l, s, t, 0, 2, i)
> > #define rwlock_acquire_read(l, s, t, i)       lock_acquire(l, s, t, 2, 2, i)
>  
> > and lockdep will allow recursive read-locking. You'll also need a 
> > lockdep_init_map() call to register the lock with lockdep. Then your 
> > locking scheme will be fully checked by lockdep too. (with your current 
> > code the new lock type is not added to the lock dependency graph(s))
> 
> I'm on it. :)

you'll also need to add a dep_map to the cpu_hotplug structure itself.

and i think this extension of lockdep to the new lock type will be 
invariant with the per-cpu optimizations i suggested in the previous 
mail: because it's only the scalability of cpu_hotplug_lock() that will 
improve [dramatically], its locking semantics wont.

	Ingo