[Devel] Re: [PATCH v7 02/10] foundations of per-cgroup memory pressure controlling.

[Glauber Costa]

On 12/04/2011 11:59 PM, KAMEZAWA Hiroyuki wrote:
> On Fri, 2 Dec 2011 15:46:46 -0200
> Glauber Costa<glommer at parallels.com>  wrote:
>
>>
>>>>    static void proto_seq_printf(struct seq_file *seq, struct proto *proto)
>>>>    {
>>>> +	struct mem_cgroup *memcg = mem_cgroup_from_task(current);
>>>> +
>>>>    	seq_printf(seq, "%-9s %4u %6d  %6ld   %-3s %6u   %-3s  %-10s "
>>>>    			"%2c %2c %2c %2c %2c %2c %2c %2c %2c %2c %2c %2c %2c %2c %2c %2c %2c %2c %2c\n",
>>>>    		   proto->name,
>>>>    		   proto->obj_size,
>>>>    		   sock_prot_inuse_get(seq_file_net(seq), proto),
>>>> -		   proto->memory_allocated != NULL ? atomic_long_read(proto->memory_allocated) : -1L,
>>>> -		   proto->memory_pressure != NULL ? *proto->memory_pressure ? "yes" : "no" : "NI",
>>>> +		   sock_prot_memory_allocated(proto, memcg),
>>>> +		   sock_prot_memory_pressure(proto, memcg),
>>>
>>> I wonder I should say NO, here. (Networking guys are ok ??)
>>>
>>> IIUC, this means there is no way to see aggregated sockstat of all system.
>>> And the result depends on the cgroup which the caller is under control.
>>>
>>> I think you should show aggregated sockstat(global + per-memcg) here and
>>> show per-memcg ones via /cgroup interface or add private_sockstat to show
>>> per cgroup summary.
>>>
>>
>> Hi Kame,
>>
>> Yes, the statistics displayed depends on which cgroup you live.
>> Also, note that the parent cgroup here is always updated (even when
>> use_hierarchy is set to 0). So it is always possible to grab global
>> statistics, by being in the root cgroup.
>>
>> For the others, I believe it to be a question of naturalization. Any
>> tool that is fetching these values is likely interested in the amount of
>> resources available/used. When you are on a cgroup, the amount of
>> resources available/used changes, so that's what you should see.
>>
>> Also brings the point of resource isolation: if you shouldn't interfere
>> with other set of process' resources, there is no reason for you to see
>> them in the first place.
>>
>> So given all that, I believe that whenever we talk about resources in a
>> cgroup, we should talk about cgroup-local ones.
>
> But you changes /proc/ information without any arguments with other guys.
> If you go this way, you should move this patch as independent add-on patch
> and discuss what this should be. For example, /proc/meminfo doesn't reflect
> memcg's information (for now). And scheduler statiscits in /proc/stat doesn't
> reflect cgroup's information.

No, I do not.
I may not have discussed it with everybody, but I did send some mails 
about it a while ago:

https://lkml.org/lkml/2011/10/3/60 (I sent it to containers as well 
once, but I now realize it was during the time the ML was down).

At the time, *I* was probably the only one, arguing not to do it. I've 
changed my mind since then.

> So, please discuss the problem in open way. This issue is not only related to
> this patch but also to other cgroups. Sneaking this kind of _big_ change in
> a middle of complicated patch series isn't good.

Absolutely. I can even remove this entirely and queue it for a following 
patchset if you prefer.

> In short, could you divide this patch into a independent patch and discuss
> again ? If we agree the general diection should go this way, other guys will
> post patches for cpu, memory, blkio, etc.

Yes I can.

I am expanding the CC list here so other people that cares for other 
controllers can chime in. You are welcome to give your opinion as the 
memcg maintainer as well.