Re: [PATCH v3] nvmet: add cgroup_path to charge namespace I/O to a cgroup

[peng yu]

On Wed, Sep 23, 2026 at 8:53 AM Michal Koutný <mkoutny@xxxxxxxx> wrote:
>
> On Wed, Sep 23, 2026 at 08:26:53AM -0700, Peng Yu <yupeng0921@xxxxxxxxx> wrote:
> > Scenario:
> > * Create multiple nvmet subsystems/namespaces.
> > * The namespaces are backed by different LVM logical volumes.
> > * Some of the logical volumes share the same physical volumes.
>
> Why are LVMs mentioned? (The test scenario doesn't seem to use those.
> And the example is backed by RAM, so where would be any IO to control at
> all? Note: I'm only giving this part of my attention span.)

The LVMs are close to my real use case, so I mentioned them to explain why
I need this feature.
When I write the testing code, I try to demonstrate the usage in a
simpler way, so
I use RAM devices.
Sorry for the inconsistency.


> > * The subsystems are exported to different users.
> > * We should provide each user a specific iops/bps quota, thus a noisy
> >   neighbor won't impact the performance of other logical volumes.
>
> Why cannot you place users into respective cgroups and configure
> appropriate per-device limits?

I export the devices to nvme target. Per my understanding, the IOs are
controlled by kernel threads. The users are remote users, so I can't find
a way to put them in a local cgroup.

> Thanks for providing more context about the scenario so that I can
> understand what's the goal and obstacle.
>
> Michal