[RFD] Merge task counter into memcg

[Frederic Weisbecker]

On Thu, Apr 12, 2012 at 09:56:49AM +0900, KAMEZAWA Hiroyuki wrote:
> (2012/04/12 3:57), Frederic Weisbecker wrote:
> 
> > Hi,
> > 
> > While talking with Tejun about targetting the cgroup task counter subsystem
> > for the next merge window, he suggested to check if this could be merged into
> > the memcg subsystem rather than creating a new one cgroup subsystem just
> > for task count limit purpose.
> > 
> > So I'm pinging you guys to seek your insight.
> > 
> > I assume not everybody in the Cc list knows what the task counter subsystem
> > is all about. So here is a summary: this is a cgroup subsystem (latest version
> > in https://lwn.net/Articles/478631/) that keeps track of the number of tasks
> > present in a cgroup. Hooks are set in task fork/exit and cgroup migration to
> > maintain this accounting visible to a special tasks.usage file. The user can
> > set a limit on the number of tasks by writing on the tasks.limit file.
> > Further forks or cgroup migration are then rejected if the limit is exceeded.
> > 
> > This feature is especially useful to protect against forkbombs in containers.
> > Or more generally to limit the resources on the number of tasks on a cgroup
> > as it involves some kernel memory allocation.
> > 
> > Now the dilemna is how to implement it?
> > 
> > 1) As a standalone subsystem, as it stands currently (https://lwn.net/Articles/478631/)
> > 
> > 2) As a feature in memcg, part of the memory.kmem.* files. This makes sense
> > because this is about kernel memory allocation limitation. We could have a
> > memory.kmem.tasks.count
> > 
> > My personal opinion is that the task counter brings some overhead: a charge
> > across the whole hierarchy at every fork, and the mirrored uncharge on task exit.
> > And this overhead happens even in the off-case (when the task counter susbsystem
> > is mounted but the limit is the default: ULLONG_MAX).
> > 
> > So if we choose the second solution, this overhead will be added unconditionally
> > to memcg.
> > But I don't expect every users of memcg will need the task counter. So perhaps
> > the overhead should be kept in its own separate subsystem.
> > 
> > OTOH memory.kmem.* interface would have be a good fit.
> > 
> > What do you think?
> 
> 
> Sounds interesting to me. Hm, does your 'overhead' of task accounting is
> enough large to be visible to users ? How performance regression is big ?

I haven't measured. But on every fork, we do a res_counter_charge() that
walks through css_set and all its css_set ancestors, take a spinlock and
increment something to every level. In terms of cache trashing and algorithm
complexity, I believe the issue is real.

> BTW, now, all memcg's limit interfaces use 'bytes' as an unit of accounting.
> It's a small concern to me to have mixture of bytes and numbers of objects
> for accounting.

Indeed, this can be confusing for users.

> But I think increasing number of subsystem is not very good....

If the result is a better granularity on the overhead, I believe this
can be a good thing.