Re: [PATCH v3 00/14] mm, swap: extendable swap devices (xswap)

[Chris Li]

On Fri, Oct 2, 2026 at 10:22 AM David Hildenbrand (Arm)
<david@xxxxxxxxxx> wrote:
>
> On 9/17/26 15:17, Johannes Weiner wrote:
> >> And the limit is only an optional knob. If the admin does not set it,
> >> the device grows to the full address space, so there is no space limit
> >> to hit at all. It already behaves the way you want by default. The knob
> >> is only for admins who want a ceiling, they can use it or not. I hope
> >> this would not be a problem for your use case.
> >
> > No, I've laid this out already as well.
> >
> > This isn't about "my" usecase. It's about designing a coherent
> > interface that works well with a large number of usecases, and other
> > pieces of kernel infrastructure commonly used in conjunction.
>
> Johannes, I read this as a NAK for now to the current approach from the memcg
> side, is that correct?

I will let Johannes comment on his take. Here is the TL;DR: from my
point of view:

Johannes does not want to allocate the vswap/xswap entry charge to memory.swap.
So he made two requests for xswap to move forward:
1) Must not charge to memory.swap
2) Make the xswap initial size large enough

Regarding the first point (the charging problem), Johannes later wrote
up the two different use cases very nicely here:

[1] https://lore.kernel.org/all/araWKyJ5MENBwXCL@xxxxxxxxxxx/

So I am not sure he is still going to NACK on it. I wish he had
outlined a path for usage case 1) in that email.

Zswap currently charges to memory.swap when allocating an entry for
zswap. So I consider it a user-visible behavior change; we should not
silently change the behavior. (Others disagree because current zswap
shares swap entries with disk as well).
I am hoping that we can find some way out for the charging usage case
1) in the emali [1] e.g. have a config option to let the user select
which charging behavior they want.

Otherwise, we will need to find an escalation path to resolve the
different opinion on this.

>
> I've tried to follow the discussion on the other thread, but it just exploded
> and some of the things I read there made me rather upset late on a Friday evening.

If you want to have peace of mind, that's the wrong thread to read
late on a Friday evening :-)

Chris