Re: memo: mem+swap controller

[Balbir Singh <balbir@linux.vnet.ibm.com>]

Daisuke Nishimura wrote:
> On Fri, 01 Aug 2008 08:58:07 +0530, Balbir Singh <balbir@linux.vnet.ibm.com> wrote:
>> Daisuke Nishimura wrote:
>>> Hi, Kamezawa-san.
>>>
>>> On Thu, 31 Jul 2008 10:15:33 +0900, KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com> wrote:
>>>> Hi, mem+swap controller is suggested by Hugh Dickins and I think it's a great
>>>> idea. Its concept is having 2 limits. (please point out if I misunderstand.)
>>>>
>>>>  - memory.limit_in_bytes       .... limit memory usage.
>>>>  - memory.total_limit_in_bytes .... limit memory+swap usage.
>>>>
>>> When I've considered more, I wonder how we can accomplish
>>> "do not use swap in this group".
>>>
>> It's easy use the memrlimit controller and set virtual address limit <=
>> memory.limit_in_bytes. I use that to make sure I never swap out.
>>
> I don't think it works under memory pressure *outside* of the group,
> that is, global memory reclaim.
> (I think "limit_in_bytes == total_limit_in_bytes" also works *inside* memory 
> pressure.)
> 

Yes, but it ensures that the current cgroup does not add to global swap
pressure. Trying to control global pressure from inside a cgroup might be hard,
but I see your intention of isolating the cgroup from global pressure. With
swappiness, it should be possible to purge page cache ahead of creating swap
pressure.

>>> Setting "limit_in_bytes == total_limit_in_bytes" doesn't meet it, I think.
>>> "limit_in_bytes = total_limit_in_bytes = 1G" cannot
>>> avoid "memory.usage = 700M swap.usage = 300M" under memory pressure
>>> outside of the group(and I think this behavior is the diffrence
>>> of "memory controller + swap controller" and "mem+swap controller").
>>>
>>> I think total_limit_in_bytes and swappiness(or some flag to indicate
>>> "do not swap out"?) for each group would make more sense.
>> I do intend to add the swappiness feature soon for control groups.
>>
> How does it work?
> Does it affect global page reclaim?
> 

We have a swappiness parameter in scan_control. Each control group indicates
what it wants it swappiness to be when the control group is over it's limit and
reclaim kicks in.

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

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>