Re: [patch] Reorganize the cpufreq cpu hotplug locking to not be totally bizare

[Ingo Molnar <mingo@elte.hu>]

* Linus Torvalds <torvalds@osdl.org> wrote:

> But I agree with Arjan - I think the fundamental problem is that cpu 
> hotplug locking is just is fundamentally badly designed as-is. There's 
> really very little point to making it a _lock_ per se, since most 
> people really want more of a "I'm using this CPU, don't try to remove 
> it right now" thing which is more of a ref-counting-like issue.

we'd also need a facility to wait on that refcount - i.e. a waitqueue. 
Which means we'd have a "refcount + waitqueue", which is equivalent to a 
"recursive, sleeping read-lock", where the write-side could be used as a 
simple facility to "wait for all readers to go away and block new 
readers from entering the critical sections". [which type of lock Linux 
does not have right now. rwsems come the closest but they dont recurse.]

Also, the hotplug lock is global right now which is pretty unscalable, 
so the rw-mutex should also be per-CPU, and the hotplug locking API 
should be changed to something like:

	cpu = cpu_hotplug_lock();
	...
	cpu_hotplug_unlock(cpu);

To enable a task to schedule away (and potentially migrate to another 
CPU) with the per-CPU lock held but still be able to unlock the right 
per-cpu lock. [this above approach is quite similar to how we do 
sleeping per-cpu locks in -rt.]

	Ingo