Re: [PATCH] mm: memcontrol: reclaim when shrinking memory.high below usage

[Michal Hocko <mhocko@kernel.org>]

On Fri 11-03-16 12:13:04, Vladimir Davydov wrote:
> On Fri, Mar 11, 2016 at 09:42:39AM +0100, Michal Hocko wrote:
> > On Fri 11-03-16 11:34:40, Vladimir Davydov wrote:
> > > On Thu, Mar 10, 2016 at 03:50:13PM -0500, Johannes Weiner wrote:
> > > > When setting memory.high below usage, nothing happens until the next
> > > > charge comes along, and then it will only reclaim its own charge and
> > > > not the now potentially huge excess of the new memory.high. This can
> > > > cause groups to stay in excess of their memory.high indefinitely.
> > > > 
> > > > To fix that, when shrinking memory.high, kick off a reclaim cycle that
> > > > goes after the delta.
> > > 
> > > I agree that we should reclaim the high excess, but I don't think it's a
> > > good idea to do it synchronously. Currently, memory.low and memory.high
> > > knobs can be easily used by a single-threaded load manager implemented
> > > in userspace, because it doesn't need to care about potential stalls
> > > caused by writes to these files. After this change it might happen that
> > > a write to memory.high would take long, seconds perhaps, so in order to
> > > react quickly to changes in other cgroups, a load manager would have to
> > > spawn a thread per each write to memory.high, which would complicate its
> > > implementation significantly.
> > 
> > Is the complication on the managing part really an issue though. Such a
> > manager would have to spawn a process/thread to change the .max already.
> 
> IMO memory.max is not something that has to be changed often. In most
> cases it will be set on container start and stay put throughout
> container lifetime. I can also imagine a case when memory.max will be
> changed for all containers when a container starts or stops, so as to
> guarantee that if <= N containers of M go mad, the system will survive.
> In any case, memory.max is reconfigured rarely, it rather belongs to the
> static configuration.

I see
 
> OTOH memory.low and memory.high are perfect to be changed dynamically,
> basing on containers' memory demand/pressure. A load manager might want
> to reconfigure these knobs say every 5 seconds. Spawning a thread per
> each container that often would look unnecessarily overcomplicated IMO.

The question however is whether we want to hide a potentially costly
operation and have it unaccounted and hidden in the kworker context.
I mean fork() + write() doesn't sound terribly complicated to me to have
a rather subtle behavior in the kernel.
-- 
Michal Hocko
SUSE Labs