Re: [PATCH v4 00/14] mm, swap: extendable swap devices (xswap phase I)

[Baoquan He <baoquan.he@linux.dev>]

On 10/12/26 at 03:44pm, David Hildenbrand (Arm) wrote:
> On 10/12/26 11:16, Baoquan He wrote:
> > Hi Klara,
> > 
> > On 10/07/26 at 06:28pm, Klara Modin wrote:
> >> On 2026-10-03 08:31:22 +0800, Baoquan He wrote:
> > ...snip...
> >>> The only argument create takes is a priority. The size comes from
> >>> xswap.max= at boot, so there is one place to ask for a size and one to
> >>> ask for a priority, and a priority can be given per device while a size
> >>> cannot:
> >>>
> >>>   # echo > /sys/kernel/mm/xswap/create          default priority
> >>>   # echo 100 > /sys/kernel/mm/xswap/create      priority 100
> >>>
> >>
> >>> The default is the highest priority, SWAP_FLAG_PRIO_MASK. An xswap
> >>> device is then the first one tried, ahead of any ordinary swap device.
> >>> Setting a lower priority is allowed.
> >>
> >> Nice, this is now a good default which the user should not have to
> >> change.
> > 
> > Thanks a lot for careful reviewing, and sorry for late reply.
> > 
> >>
> >>>
> >>> The device does not move once it exists, so the address space cannot be
> >>> changed from this interface either. xswap.max= at boot and this together
> >>> are the whole configuration.
> > ...snip...
> >>> Open items
> >>> ----------
> >>>
> >>> - xswap_create() sets si->pages to the whole address space (twice RAM by
> >>>   default) and _enable_swap_info() adds it to nr_swap_pages and
> >>>   total_swap_pages, which is what __vm_enough_memory() uses for
> >>>   overcommit. That raises the commit limit by 2xRAM for a device whose
> >>>   real capacity is bounded by the zswap pool and by how well the workload
> >>>   compresses. The question is which semantics overcommit should follow:
> >>>   worst-case backing, or the address space.
> >>>
> >>> - destroy takes a bare swap type. The type is an internal index that
> >>>   alloc_swap_info() recycles as soon as SWP_USED clears, so a value read
> >>>   earlier can name a different device by the time it is written. The
> >>>   swapoff ABI uses a path for this reason. It wants a generation or
> >>>   another stable handle.
> >>>
> >>> - The new interfaces (/sys/kernel/mm/xswap/create and destroy, xswap.max=,
> >>>   the new type string in /proc/swaps) are not documented yet.
> >>
> >> To me it feels a bit weird to have the interface split between a kernel
> >> parameter and sysfs, meaning I have to configure this in two places now.
> >> As far as I gather, there's also nothing that needs the limit to be
> >> fixed at boot since it is applied at creation. If we want to expose this
> >> limit to users (as xswap does) it would be nice to have a 'max'
> >> shorthand for tha largest possible value, i.e. SWAPFILE_CLUSTER *
> >> UNIT_MAX (or whatever it might be in the future). That would I guess in
> >> practice give the same behaviour as vm.overcommit_memory=1 which might
> >> or might not be desirable.
> > 
> > Thanks for the comment.
> > 
> > Here xswap.max= is a global boot parameter, while not a per-device or
> > runtie knob. It sets the space space (si->max) of the xswap device
> > at boot and can't change at runtime. So on the two places: the size is
> > a global setting, fixed at boot; the priority is per device, it is in
> > the create interface. They are different things.
> > 
> > The default is 2xRAM. From our earlier discussion, this is the largest
> > size we think is sensible, and in practice I don't think it is not
> > reachable in reality. If a system has swapped out 70% of its memory, and
> > even more at 100%, it is already doubtful that it can still function
> > normally. So 2xRAM is an upper bound, not a target.
> 
> I remember some overcommitted hypervisor setups where people swapped out
> significantly more -- without any issues.
> 
> For example, Windows VMs initialize all VM memory when booting up. Afterwards,
> you might just have plenty of memory that was touched once but will not really
> be used in any near future.
> 
> So I would consider some crazy hypervisor scenarios as well where people
> commonly have significant cold memory that is currently not hot (not in the
> working set) but might possibly used in some distant future.

Hi David,

Thanks a lot for the valuable input. This is probably the first first-hand
information I have heard about a production environment. And Klara
measured about 160% of RAM in zswap on her RISC-V box, so the 2xRAM default
is reachable in practice, not just in theory.

So I see 2xRAM as a default, not as a cap. xswap.max= already overrides it.
After your comment and Klara's numbers, I am open to a larger default, or to
a "max" value for the address space. The real limit is still the backing:
the zswap pool plus the block devices. The address space should no longer
be the limiting factor.

By the way, in Windows-VM case you mentioned, is there a block swap backend,
or is it zswap alone? If it is zswap alone, the knob that matters is
zswap.max_pool_percent, because it sets how much cold memory the pool can
hold.

Thanks
Baoquan