Re: [RFC PATCH v2 0/7] mm, swap: Virtual Swap Space (Swap Table Edition)

[YoungJun Park]

...
> * Integration with swap.tier by Youngjun (see [12]). For now, I'm
>   leaning towards opting out the vswap device from swap.tier entirely, and
>   treat it as a special device. Integrating it with swap.tiers will
>   benefit the cases where you want some cgroups to skip vswap for fast
>   swap devices (pmem), whereas other should go through zswap first. But
>   most other use cases, either the overhead of vswap will be acceptable
>   (or not the bottleneck), or we can just disable CONFIG_VSWAP entirely :)
> 
>   Youngjun, may I ask for your thoughts on this?

Hi Nhat,

Tier 1: VSWAP, Tier 2: ZSWAP ...

I don't see any problem applying the desired functionality with the
currently proposed mechanism and interface. With this, a user would be
assigned the default Virtual -> RAM swap tier, and the overall picture
becomes one where swap tiers are composed according to the priority
setting.

A few more thoughts came to mind.

Shakeel also proposed a per-tier max for the swap tier interface.

https://lore.kernel.org/linux-mm/aiw2p5ANjsQUCIHA@xxxxxxxxx/

However, for vswap, rather than treating it as a case for limiting the
amount via such a per-tier max, I think the current interface is the
better fit. (But, as Shakeel mentioned, if we only allow the limit
to be set to 0 or max, the usage could end up being the same. I'm still
thinking this part through.)

I have a few other thoughts as well, but I plan to raise those points in
the swap tier discussion thread instead. Please take a look at the
related thread, and let me know if you have any opinions. :)

And I'll share more if other thoughts come to mind

Thanks,
Youngjun Park