Re: [PATCH RFC 2/2] cgroup/dmem: add a node to double charge in memcg

[Eric Chanudet <echanude@redhat.com>]

On Tue, Apr 07, 2026 at 02:48:05PM +0200, Michal Koutný wrote:
> Hi.
> 
> On Fri, Apr 03, 2026 at 10:08:36AM -0400, Eric Chanudet <echanude@redhat.com> wrote:
> > Introduce /cgroupfs/<>/dmem.memcg to make allocations in a dmem
> > controlled region also be charged in memcg.
> > 
> > This is disabled by default and requires the administrator to configure
> > it through the cgroupfs before the first charge occurs.
> 
> This somehow dropped the reason from [1] that this should be per-cgroup
> controllable. Is that still valid? (Otherwise, I'd ask why not make this
> a simple boot cmdline parameter like cgroup.memory=nokmem.)

[1] argued it should be controllable per dmem region more than per
cgroup. For example, a cgroup configured +memory and +dmem with one
region charging only dmem and the other double charging memcg and dmem.

cgroup.memory=nokmem karg would double charge all or none of the regions
for all cgroups iiuc. So maybe just make memcg CFTYPE_ONLY_ON_ROOT and
inherit that configuration in all children would do?

> > @@ -624,6 +656,13 @@ void dmem_cgroup_uncharge(struct dmem_cgroup_pool_state *pool, u64 size)
> >  		return;
> >  
> >  	page_counter_uncharge(&pool->cnt, size);
> > +
> > +	struct mem_cgroup *memcg = mem_cgroup_from_cgroup(pool->cs->css.cgroup);
> 
> This is not necessarily same memcg as when the dmem was charged via
> current (imagine dmem controller to depth N, but memcg only to N-1;
> charge, then memcg is enabled up to N so this would attempt uncharge
> from new memcg at level N, possibly going negative).
> 
> There is a question whether dmem should enforce same-depth hierarchies
> with `dmem_cgrp_subsys.depends_on = 1 << memory_cgrp_id` (see
> io_cgrp_subsys for comparison).

Thanks! I'll look into depends_on.

> And eventually, if per-cgroup attribute is desired, it would make
> greater sense to me if that attribute was on the parent level, so that
> siblings competing among each other are always of the same composition
> (i.e. all w/out dmem or all including dmem). This likely results in this
> extra-charging attribute to be properly hierarchical.
> 
> HTH,
> Michal
> 
> [1] https://lore.kernel.org/all/a446b598-5041-450b-aaa9-3c39a09ff6a0@amd.com/

Thank you for the feedback,

-- 
Eric Chanudet