On Thu, Jan 24, 2019 at 09:22:52AM +0100, Michal Hocko wrote: > On Wed 23-01-19 17:31:44, Chris Down wrote: > > Acked-by: Johannes Weiner <hannes@xxxxxxxxxxx> > > btw. I do not see this patch posted anywhere yet it already comes with > an ack. Have I just missed a previous version? I reviewed it offline before Chris sent it out. I agree with the sentiment that the non-hierarchical behavior was an oversight, not a design decision. The arguments against the change don't convince me: the added difficulty of finding out local values is true for all other cgroup files as well. This is traded off with being able to detect any subtree state from the first level cgroups and drill down on-demand, without having to scan the entire tree on each monitoring interval. That's a trade-off we've made everywhere else, so this is simply an inconsistency, not a legitimate exception to the rule. We cannot fully eliminate a risk for regression, but it strikes me as highly unlikely, given the extremely young age of cgroup2-based system management and surrounding tooling. Acked-by: Johannes Weiner <hannes@xxxxxxxxxxx>
