Re: [PATCH v3 22/23] xen/Kconfig: introduce HAS_STATIC_MEMORY

[Oleksii Kurochko <oleksii.kurochko@xxxxxxxxx>]

On 6/23/26 10:26 AM, Jan Beulich wrote:
> On 17.06.2026 13:17, Oleksii Kurochko wrote:
> > Architectures that implement guest_physmap_add_pages() select
> > HAS_STATIC_MEMORY; STATIC_MEMORY then depends on it.  ARM selects the
> > new flag; RISC-V does not, so CONFIG_STATIC_MEMORY is unavailable there
> > and randconfig builds no longer require an explicit STATIC_MEMORY=n
> > override to avoid a compilation error.
> How did you come up with the connection to guest_physmap_add_pages()?
It is because of you mentioned in this sentense ...

> That's a close sibling of guest_physmap_add_page(), and they all should
> fall in the same group. The fact that right now static-mem is the only
> caller of guest_physmap_add_pages() is secondary.
... (the last sentence)

> New callers could
> appear. guest_physmap_add_page() could likely (in principle) be
> implemented in terms of guest_physmap_add_pages().
>
> What you're after is a way to {en,dis}able STATIC_MEMORY on a per-arch
> basis. That's all what matters here.
I will reword that part in the following way:

Introduce HAS_STATIC_MEMORY so that STATIC_MEMORY can be enabled or
disabled on a per-architecture basis. An architecture that supports
static memory selects HAS_STATIC_MEMORY, and STATIC_MEMORY depends on
it. ARM selects the new flag; RISC-V does not, so CONFIG_STATIC_MEMORY
is unavailable on RISC-V and randconfig builds no longer require an
explicit STATIC_MEMORY=n override to avoid a compilation error.

> > --- a/xen/common/Kconfig
> > +++ b/xen/common/Kconfig
> > @@ -161,6 +161,9 @@ config HAS_SCHED_GRANULARITY
> >   config HAS_SHARED_INFO
> >         bool
> >   
> > +config HAS_STATIC_MEMORY
> > +       bool
> > +
> >   config HAS_SOFT_RESET
> >         bool
> >   
> > @@ -196,7 +199,7 @@ config NUMA
> >   
> >   config STATIC_MEMORY
> >         bool "Static Allocation Support (UNSUPPORTED)" if UNSUPPORTED
> > -       depends on DOM0LESS_BOOT && HAS_DEVICE_TREE_DISCOVERY
> > +       depends on HAS_STATIC_MEMORY && DOM0LESS_BOOT && 
> > HAS_DEVICE_TREE_DISCOVERY
> This may end up slightly neater as
>
>         depends on HAS_STATIC_MEMORY
>         depends on DOM0LESS_BOOT && HAS_DEVICE_TREE_DISCOVERY
I am okay with your suggestion and will apply it.

Thanks.

~ Oleksii