Re: [PATCH v5 0/4] cgroup: dmem: add selftest helper, coverage, and VM runner

[Albert Esteve]

On Mon, Aug 24, 2026 at 8:27 PM Michal Koutný <mkoutny@xxxxxxxx> wrote:
>
> Hello Albert.
>
> On Mon, Jul 06, 2026 at 02:06:39PM +0200, Albert Esteve <aesteve@xxxxxxxxxx> wrote:
> > Hi all,
> >
> > This small series adds practical test coverage for the dmem
> > cgroup controller.
> >
> > The motivation came from following the recent dmem API discussion in
> > thread [1]. That discussion considered changing the dmem API and
> > adding a new knob. Currently there are no dedicated tests covering
> > dmem behaviour, which makes such changes riskier.
> >
> > Adding selftests has an additional challenge: dmem charging paths
> > are driver-driven today, so regression testing is harder unless a
> > suitable driver is present in the test environment.
> >
> > This series addresses that by adding:
> > - a kernel-side selftest helper module to trigger charge/uncharge
> >   from userspace in a controlled way,
> > - cgroup selftests covering dmem accounting and protection semantics
> >   (including dmem.max enforcement and byte-granularity checks),
> > - a virtme-based VM runner for repeatable execution of the dmem tests.
> >
> > The goal is to make dmem behavior easier to validate when evolving the API
> > and implementation, while keeping tests deterministic and driver-independent.
>
> Thanks for the dmem selftesting. I'll respond to first two patches
> individually. I find the wrapper scripts in the last two a bit of an
> overkill for the selftest (as it adds dependency on particular VM
> environment, I'd keep the simplicity of selftests such that wherever
> they are executed (w/out wrappers), there they run (be it a physical
> machine or a VM). At most, I'd consider a simple wrapper that'd load
> ensure the test module is loaded.

Thanks for the review!

The last two patches were meant to follow the existing virtme-ng
vmtest.sh pattern used by other selftests, mainly as a convenient
developer loop. Anyway, I can see how it might be overkill, especially
in this initial version. I'll drop them for now and keep the first two
patches.

If a tiny wrapper that only ensures dmem_selftest is loaded turns out
to be useful, that can be added later.

BR,
Albert.

>
> Thanks,
> Michal