Re: [PATCH v4 08/14] acpi/generic_event_device: add logic to detect if HEST addr is available

[Igor Mammedov]

On Thu, 27 Feb 2025 08:26:38 +0100
Mauro Carvalho Chehab <mchehab+huawei@xxxxxxxxxx> wrote:

> Em Thu, 27 Feb 2025 08:19:27 +0100
> Mauro Carvalho Chehab <mchehab+huawei@xxxxxxxxxx> escreveu:
> 
> > Em Wed, 26 Feb 2025 16:52:26 +0100
> > Igor Mammedov <imammedo@xxxxxxxxxx> escreveu:
> >   
> > > On Fri, 21 Feb 2025 15:35:17 +0100
> > > Mauro Carvalho Chehab <mchehab+huawei@xxxxxxxxxx> wrote:
> > >     
> >   
> > > > diff --git a/hw/acpi/generic_event_device.c b/hw/acpi/generic_event_device.c
> > > > index 5346cae573b7..14d8513a5440 100644
> > > > --- a/hw/acpi/generic_event_device.c
> > > > +++ b/hw/acpi/generic_event_device.c
> > > > @@ -318,6 +318,7 @@ static void acpi_ged_send_event(AcpiDeviceIf *adev, AcpiEventStatusBits ev)
> > > >  
> > > >  static const Property acpi_ged_properties[] = {
> > > >      DEFINE_PROP_UINT32("ged-event", AcpiGedState, ged_event_bitmap, 0),
> > > > +    DEFINE_PROP_BOOL("x-has-hest-addr", AcpiGedState, ghes_state.use_hest_addr, false),      
> > > 
> > > you below set it for 9.2 to false, so
> > > shouldn't it be set to true by default here?    
> > 
> > Yes, but it is too early to do that here, as the DSDT table was not
> > updated to contain the GED device.
> > 
> > We're switching it to true later on, at patch 11::
> > 
> > 	d8c44ee13fbe ("arm/virt: Wire up a GED error device for ACPI / GHES")  

After sleeping on it,
what you did here is totally correct.

You are right, We can't really flip switch to true here
since without 11/14 APEI will stop working properly.

Perhaps add to commit message a note explaining why it's false
in this patch and where it will be set to true.

> 
> Hmm... too many rebases that on my head things are becoming shady ;-)
> 
> Originally, this was setting it to true, but you requested to move it
> to another patch during one of the patch reorder requests.
> 
> Anyway, after all those rebases, I guess it is now safe to set it
> to true here without breaking bisectability. I'll move the hunk back
> to this patch.
> 
> Thanks,
> Mauro
>