Re: [PATCH] mm/vmscan: respect cpuset policy during page demotion

[Michal Hocko <mhocko@suse.com>]

On Mon 07-11-22 16:05:37, Feng Tang wrote:
> On Mon, Oct 31, 2022 at 03:32:34PM +0100, Michal Hocko wrote:
> > > > OK, then let's stop any complicated solution right here then. Let's
> > > > start simple with a per-mm flag to disable demotion of an address space.
> > > > Should there ever be a real demand for a more fine grained solution
> > > > let's go further but I do not think we want a half baked solution
> > > > without real usecases.
> > > 
> > > Yes, the concern about the high cost for mempolicy from you and Yang is
> > > valid. 
> > > 
> > > How about the cpuset part?
> > 
> > Cpusets fall into the same bucket as per task mempolicies wrt costs. Geting a
> > cpuset requires knowing all tasks associated with a page. Or am I just
> > missing any magic? And no memcg->cpuset association is not a proper
> > solution at all.
> 
> No, you are not missing anything. It's really difficult to find a
> solution for all holes. And the patch is actually a best-efforts
> approach, trying to cover cgroup v2 + memory controller enabled case,
> which we think is a common user case for newer platforms with tiering
> memory.

Best effort is OK but it shouldn't create an unexpected behavior and
this approach does that.

I thought I have already explained that. But let me be more
explicit this time.  Please have a look at how controllers can be
enabled/disabled at different levels of the hierarchy. Unless memcg
grows a hard dependency on another controller (as it does with the blk
io controller) then this approach can point to a wrong cpuset. See my
point?

Really, solution for this is not going to be cheap and also I am not
sure all the hessles is really worth it until there is a clear usecase
in sight.
-- 
Michal Hocko
SUSE Labs