Re: [PATCH v3] ACPI: scan: Avoid registering platform devices with resource overlaps

On Wed, Aug 19, 2026 at 7:32 PM Nathan Chancellor <[email protected]> wrote:
>
> On Wed, Aug 19, 2026 at 03:55:43PM +0200, Rafael J. Wysocki wrote:
> > And below is one to try.
> >
> > ---
> >  drivers/acpi/acpi_platform.c |    7 ++++++-
> >  1 file changed, 6 insertions(+), 1 deletion(-)
> >
> > --- a/drivers/acpi/acpi_platform.c
> > +++ b/drivers/acpi/acpi_platform.c
> > @@ -85,7 +85,12 @@ static unsigned int acpi_platform_adjust
> >       for (i = 0; i < count; ) {
> >               struct resource *res = &resources[i];
> >
> > -             if (resource_type(new_res) != resource_type(res) ||
> > +             /*
> > +              * Look for overlaps of resources of the same type that
> > +              * would cause resource insertion to fail.
> > +              */
> > +             if (__resource_contains_unbound(new_res, res) ||
> > +                 resource_type(new_res) != resource_type(res) ||
> >                   !resource_union(new_res, res, new_res)) {
> >                       i++;
> >                       continue;
> >
> >
> >
>
> This does not appear to resolve the probe failure for me.

So can you please try to reverse new_res and res in the
__resource_contains_unbound() check in the patch and try again?

> Attached are the output of dmesg from a boot with that patch (let me know if you want
> a clean one) and the output of acpidump.

I'll get to the acpidump next, but I'm not sure if I can do it today.