Re: [PATCH v1 3/3] mm: per-process reclaim

[Minchan Kim]

Hi Johannes,

On Mon, Jun 13, 2016 at 11:06:53AM -0400, Johannes Weiner wrote:
> Hi Minchan,
> 
> On Mon, Jun 13, 2016 at 04:50:58PM +0900, Minchan Kim wrote:
> > These day, there are many platforms available in the embedded market
> > and sometime, they has more hints about workingset than kernel so
> > they want to involve memory management more heavily like android's
> > lowmemory killer and ashmem or user-daemon with lowmemory notifier.
> > 
> > This patch adds add new method for userspace to manage memory
> > efficiently via knob "/proc/<pid>/reclaim" so platform can reclaim
> > any process anytime.
> 
> Cgroups are our canonical way to control system resources on a per
> process or group-of-processes level. I don't like the idea of adding
> ad-hoc interfaces for single-use cases like this.
> 
> For this particular case, you can already stick each app into its own
> cgroup and use memory.force_empty to target-reclaim them.
> 
> Or better yet, set the soft limits / memory.low to guide physical
> memory pressure, once it actually occurs, toward the least-important
> apps? We usually prefer doing work on-demand rather than proactively.
> 
> The one-cgroup-per-app model would give Android much more control and
> would also remove a *lot* of overhead during task switches, see this:
> https://lkml.org/lkml/2014/12/19/358

I didn't notice that. Thanks for the pointing.
I read the thread you pointed out and read memcg code.

Firstly, I thought one-cgroup-per-app model is abuse of memcg but now
I feel your suggestion does make sense that it's right direction for
control memory from the userspace. Just a concern is that not sure
how hard we can map memory management model from global memory pressure
to per-app pressure model smoothly.

A question is it seems cgroup2 doesn't have per-cgroup swappiness.
Why?

I think we need it in one-cgroup-per-app model.