Re: [External] Re: [RFC 2/5] memcontrol: add boot option to enable memsw account on dfl

[jingxiang zeng <jingxiangzeng.cas@gmail.com>]

On Thu, 3 Apr 2025 at 15:47, Zhongkun He <hezhongkun.hzk@bytedance.com> wrote:
>
> On Wed, Apr 2, 2025 at 9:42 PM Michal Koutný <mkoutny@suse.com> wrote:
> >
> > On Thu, Mar 20, 2025 at 08:33:09AM -0700, Shakeel Butt <shakeel.butt@linux.dev> wrote:
> > > However I want us to discuss and decide the semantics of
> > > memsw from scratch rather than adopting v1 semantics.
> >
> > +1
> >
> > > Also we should discuss how memsw and swap limits would interact and
> > > what would be the appropriate default.
> >
> > Besides more complicated implementation, merged memsw won't represent an
> > actual resource.
> >
> > So I'd be interested in use cases (other than "used to it from v1") that
> > cannot be controlled with separate memory. and swap. limits.
> >
>
> Hi Michal
>
> We encountered an issue, which is also a real use case. With memory offloading,
> we can move some cold pages to swap. Suppose an application’s peak memory
> usage at certain times is 10GB, while at other times, it exists in a
> combination of
> memory and swap. If we set limits on memory or swap separately, it would lack
> flexibility—sometimes it needs 1GB memory + 9GB swap, sometimes 5GB
> memory + 5GB swap, or even 10GB memory + 0GB swap. Therefore, we strongly
> hope to use the mem+swap charging method in cgroupv2
>
> >
> > 0.02€,
> > Michal

Yes, in the container scenario, if swap is enabled on the server and
the customer's
container requires 10GB of memory, we only need to set
memory.memsw.limit_in_bytes=10GB, and the kernel can automatically swap out
part of the business container's memory to swap according to the server's memory
pressure, and it can be fully guaranteed that the customer's container
will not use
more memory because swap is enabled on the server.
>