On Tue, Dec 17, 2019 at 4:27 AM Giuseppe Scrivano <gscrivan@xxxxxxxxxx> wrote: > > Mina Almasry <almasrymina@xxxxxxxxxx> writes: > > > On Mon, Dec 16, 2019 at 1:27 PM Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> wrote: > >> > >> On Mon, 16 Dec 2019 12:43:48 -0800 Tejun Heo <tj@xxxxxxxxxx> wrote: > >> > >> > On Mon, Dec 16, 2019 at 08:38:31PM +0100, Giuseppe Scrivano wrote: > >> > > In the effort of supporting cgroups v2 into Kubernetes, I stumped on > >> > > the lack of the hugetlb controller. > >> > > > >> > > When the controller is enabled, it exposes four new files for each > >> > > hugetlb size on non-root cgroups: > >> > > > >> > > - hugetlb.<hugepagesize>.current > >> > > - hugetlb.<hugepagesize>.max > >> > > - hugetlb.<hugepagesize>.events > >> > > - hugetlb.<hugepagesize>.events.local > >> > > > >> > > The differences with the legacy hierarchy are in the file names and > >> > > using the value "max" instead of "-1" to disable a limit. > >> > > > >> > > The file .limit_in_bytes is renamed to .max. > >> > > > >> > > The file .usage_in_bytes is renamed to .current. > >> > > > >> > > .failcnt is not provided as a single file anymore, but its value can > >> > > be read through the new flat-keyed files .events and .events.local, > >> > > through the "max" key. > >> > > > >> > > Signed-off-by: Giuseppe Scrivano <gscrivan@xxxxxxxxxx> > >> > > >> > Acked-by: Tejun Heo <tj@xxxxxxxxxx> > >> > > >> > This can go through either the mm tree or the cgroup tree. If Andrew > >> > doesn't pick it up in several days, I'll apply it to cgroup/for-5.6. > >> > > >> > >> Thanks, I grabbed it. > >> > >> Giuseppe, yuo presumably have test code lying around. Do you have > >> something which can be tossed together for tools/testing/selftests/? > >> Presumably under cgroup/. > >> > >> We don't seem to have much in the way of selftest code for cgroups. I > >> wonder why. > > > > Just FYI I have a patch series in review that does a hefty bit of > > modifications to hugetlb_cgroup, and that comes with a decent bit of > > tests for hugetlb cgroup (and only hugetlb cgroups, I'm not looking > > into memcg tests or cgroup tests in general): > > https://lkml.org/lkml/2019/10/29/1203 > > > > If Giuseppe adds tests for hugetlb cgroup v2 that would be great, but > > if not, a decent bit of hugetlb cgroup tests should be coming your way > > as my series gets reviewed. > > I've some code I've used to test the hugetlb cgroup that I can clean up > and include in the selftests. > > Mina, are you going to rebase your patchset? I can add new tests for > cgroup v2 on top of yours. > Yep, I'm working on that now and addressing comments from my last iteration. I should be able to send it out for review before the end of the week. > Regards, > Giuseppe >
