Tejun Heo wrote on 2021/8/25 3:08: > Hello, > > On Tue, Aug 24, 2021 at 06:44:23PM +0200, Michal Koutný wrote: >> However, the non-hierarchical failcnt interface looks like v1ism to me >> (I think new features should come with v2 first in mind). >> What about exposing this in misc.events file with max.$res_name entries? > > Ah yeah, good point. misc.events sounds like a good spot to put these. > >> Or if the hierarchical reporting is unnecessary now, there can be just >> misc.events.local for starters. > > I'd prefer to stick with hierarchical counting as the first step at least. > >> (That reminds me the forgotten pids.events[.local] rework [1], oops.) >> >> https://lore.kernel.org/lkml/20191128172612.10259-1-mkoutny@xxxxxxxx/#t The core logic of pids cgroup and misc cgroup is similar. Is it possible for us to merge pids cgroup into misc cgroup? > I think both counters are useful - the number of failures due to this type > of limit in this subhierarchy, and the number of failures caused by this > particular limit in this subhierarchy. It's a pretty subtle difference to > encapsulate in a counter name tho. > > Thanks. >
