Containers: css_put() dilemma

[Balbir Singh]

Paul (??) Menage wrote:
> On 7/17/07, Balbir Singh <balbir at linux.vnet.ibm.com> wrote:
>> without too much knowledge of each other. BTW, what are the semantics
>> of css_put() is it expected to free the container/run the release agent
>> when the reference count of the container_subsys_state drops to zero?
>>
> 
> If you css_put() the last reference on a subsystem state object and
> the associated container is marked as notify_on_release, then
> check_for_release() is called which does a more full check of whether
> the container is releasable. If it is, a workqueue task is scheduled
> to run the userspace release agent, which can then do anything it
> wants, including potentially deleting the empty container.
> 

Ok.. so my problem still remains, how do I get a non-blocking atomic
reference increment/decrement routine, that would prevent my
container from being deleted?

I don't find cpusets using css_put(). I was hoping that we could
alter css_* would provide the functionality I need.


-- 
	Warm Regards,
	Balbir Singh
	Linux Technology Center
	IBM, ISTL