OpenVZ Forum: Devel » [PATCH v3 04/13] kmem accounting basic infrastructure

Home » Mailing lists » Devel » [PATCH v3 04/13] kmem accounting basic infrastructure

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Re: [PATCH v3 04/13] kmem accounting basic infrastructure [message #48082 is a reply to message #48081]

Wed, 26 September 2012 23:08

Tejun Heo
Messages: 184
Registered: November 2006

Senior Member

Hello, Glauber.

On Thu, Sep 27, 2012 at 02:54:11AM +0400, Glauber Costa wrote:
> I don't. Much has been said in the past about the problem of sharing. A
> lot of the kernel objects are shared by nature, this is pretty much
> unavoidable. The answer we have been giving to this inquiry, is that the
> workloads (us) interested in kmem accounted tend to be quite local in
> their file accesses (and other kernel objects as well).
>
> It should be obvious that not all workloads are like this, and some of
> them would actually prefer to have their umem limited only.
>
> There is nothing unreasonable in tracking user memory only.
>
> If we have a global switch for "tracking all kernel memory", who would
> you account the objects that are heavily shared to? I solve this by not
> tracking kernel memory for cgroups in such workloads. What do you propose?

One of the things wrong with that is that it exposes the limitation of
the current implementation as interface to userland, which is never a
good idea. In addition, how is userland supposed to know which
workload is shared kmem heavy or not? Details like that are not even
inherent to workloads. It's highly dependent on kernel implementation
which may change any day. If we hit workloads like that the right
thing to do is improving kmemcg so that such problems don't occur, not
exposing another switch.

If we can't make that work in reasonable (doesn't have to be perfect)
way, we might as well just give up on kmem controller. If userland
has to second-guess kernel implementation details to make use of it,
it's useless.

> > Well, that's really playing with words. Limit is per cgroup and
> > before the limit is set for the first time, everything is accounted to
> > something else. How is that keeping track?
> >
>
> Even after the limit is set, it is set only by workloads that want kmem
> to be tracked. If you want to track it during the whole lifetime of the
> cgroup, you switch it before you put tasks to it. What is so crazy about it?

The fact that the numbers don't really mean what they apparently
should mean.

> > The proposed behavior seems really crazy to me. Do people really
> > think this is a good idea?
>
> It is really sad that you lost the opportunity to say that in a room
> full of mm developers that could add to this discussion in real time,
> when after an explanation about this was given, Mel asked if anyone
> would have any objections to this.

Sure, conferences are useful for building consensus but that's the
extent of it. Sorry that I didn't realize the implications then but
conferences don't really add any finality to decisions.

So, this seems properly crazy to me at the similar level of
use_hierarchy fiasco. I'm gonna NACK on this.

Thanks.

--
tejun

Report message to a moderator

[Message index]

		[PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Tue, 18 September 2012 14:04
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Fri, 21 September 2012 16:34
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Mon, 24 September 2012 08:09
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Michal Hocko on Wed, 26 September 2012 14:03
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Wed, 26 September 2012 14:33
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Michal Hocko on Wed, 26 September 2012 16:01
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Wed, 26 September 2012 17:34
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Wed, 26 September 2012 16:36
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Wed, 26 September 2012 17:36
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Wed, 26 September 2012 17:44
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Wed, 26 September 2012 17:53
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Wed, 26 September 2012 18:01
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Wed, 26 September 2012 18:56
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Wed, 26 September 2012 19:34
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Wed, 26 September 2012 19:46
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Wed, 26 September 2012 19:56
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Wed, 26 September 2012 20:02
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Wed, 26 September 2012 20:16
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Wed, 26 September 2012 21:24
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Wed, 26 September 2012 22:10
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Wed, 26 September 2012 22:29
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Wed, 26 September 2012 22:42
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Wed, 26 September 2012 22:54
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Wed, 26 September 2012 23:08
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Wed, 26 September 2012 23:20
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Wed, 26 September 2012 23:33
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Michal Hocko on Thu, 27 September 2012 12:15
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Thu, 27 September 2012 12:20
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Michal Hocko on Thu, 27 September 2012 12:40
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Thu, 27 September 2012 12:40
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Michal Hocko on Thu, 27 September 2012 12:54
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Mel Gorman on Thu, 27 September 2012 14:28
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Thu, 27 September 2012 14:49
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Thu, 27 September 2012 14:57
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Thu, 27 September 2012 17:46
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Michal Hocko on Thu, 27 September 2012 17:56
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Thu, 27 September 2012 18:45
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Sun, 30 September 2012 07:57
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Sun, 30 September 2012 08:02
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: James Bottomley on Sun, 30 September 2012 08:56
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Sun, 30 September 2012 10:37
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: James Bottomley on Sun, 30 September 2012 11:25
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Mon, 01 October 2012 00:57
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Mon, 01 October 2012 08:43
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Mon, 01 October 2012 08:46
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Wed, 03 October 2012 22:59
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Mon, 01 October 2012 08:36
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Michal Hocko on Thu, 27 September 2012 12:08
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Thu, 27 September 2012 12:11
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Thu, 27 September 2012 14:33
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Mel Gorman on Thu, 27 September 2012 14:43
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Thu, 27 September 2012 14:58
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Thu, 27 September 2012 18:30
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Sun, 30 September 2012 08:23
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Mon, 01 October 2012 08:45
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Wed, 03 October 2012 22:54
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Thu, 04 October 2012 11:55
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Sat, 06 October 2012 02:19
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Michal Hocko on Thu, 27 September 2012 15:09
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Sun, 30 September 2012 08:47
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Michal Hocko on Mon, 01 October 2012 09:27
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Tejun Heo on Wed, 03 October 2012 22:43
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Michal Hocko on Fri, 05 October 2012 13:47
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Johannes Weiner on Wed, 26 September 2012 22:11
		Re: [PATCH v3 04/13] kmem accounting basic infrastructure By: Glauber Costa on Wed, 26 September 2012 22:45

Previous Topic:	[PATCH v3 00/16] slab accounting for memcg
Next Topic:	[RFC PATCH 0/5] net: socket bind to file descriptor introduced

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Sun Aug 31 08:12:31 GMT 2025

Total time taken to generate the page: 0.12192 seconds