Re: [PATCH 3/4] memcg: hierarchical reclaim by CSS ID

[Li Zefan <lizf@cn.fujitsu.com>]

>>>>> +	while (!ret) {
>>>>> +		rcu_read_lock();
>>>>> +		nextid = root_mem->last_scanned_child + 1;
>>>>> +		css = css_get_next(&mem_cgroup_subsys, nextid, &root_mem->css,
>>>>> +				   &found);
>>>>> +		if (css && css_is_populated(css) && css_tryget(css))
>>>> I don't see why you need to check css_is_populated(css) ?
>>>>
>>> Main reason is for sanity. I don't like to hold css->refcnt of not populated css.
>> I think this is a rare case. It's just a very short period when a cgroup is
>> being created but not yet fully created.
>>
>>> Second reason is for avoinding unnecessary calls to try_to_free_pages(),
>>> it's heavy. I should also add mem->res.usage == 0 case for skipping but not yet.
>>>
>> And if mem->res.usage == 0 is checked, css_is_popuated() is just redundant.
>>
> Hmm ? Can I check mem->res.usage before css_tryget() ?
> 

I think you can. If css != NULL, css is valid (otherwise how can we access css->flags
in css_tryget), so mem is valid. Correct me if I'm wrong. :)