Re: [PATCH] ACPI: Avoid infinite recursion when dump-vmstate

[Igor Mammedov]

On Wed, 11 Nov 2020 14:01:12 +0000
"Dr. David Alan Gilbert" <dgilbert@redhat.com> wrote:

> Is someone taking a fix for this in 5.2 - it's breaking vmstate
> comparison.
can you merge it via migration tree?

[...]

for fixed up version below
Acked-by: Igor Mammedov <imammedo@redhat.com>

> > 
> > Do you mean that we need another VMStateDescription to describe
> > AcpiGhesState instead of using VMSTATE_UINT64 directly?  Maybe like this:
> > 
> > diff --git a/hw/acpi/generic_event_device.c b/hw/acpi/generic_event_device.c
> > index 6df400e1ee16..5454be67d5f0 100644
> > --- a/hw/acpi/generic_event_device.c
> > +++ b/hw/acpi/generic_event_device.c
> > @@ -322,6 +322,16 @@ static const VMStateDescription vmstate_ged_state = {
> >      }
> >  };
> > 
> > +static const VMStateDescription vmstate_ghes = {
> > +    .name = "acpi-ghes",
> > +    .version_id = 1,
> > +    .minimum_version_id = 1,
> > +    .fields     = (VMStateField[]) {
> > +        VMSTATE_UINT64(ghes_addr_le, AcpiGhesState),
> > +        VMSTATE_END_OF_LIST()
> > +    },
> > +};
> > +
> >  static bool ghes_needed(void *opaque)
> >  {
> >      AcpiGedState *s = opaque;
> > @@ -335,7 +345,7 @@ static const VMStateDescription vmstate_ghes_state = {
> >      .needed = ghes_needed,
> >      .fields      = (VMStateField[]) {
> >          VMSTATE_STRUCT(ghes_state, AcpiGedState, 1,
> > -                       vmstate_ghes_state, AcpiGhesState),
> > +                       vmstate_ghes, AcpiGhesState),
> >          VMSTATE_END_OF_LIST()
> >      }
> >  };
> > 
> > -- 
> > Thanks,
> > Peng
> >