Re: [PATCH v4 02/14] xen/asm-generic: introduce generic device.h

[Oleksii <oleksii.kurochko@xxxxxxxxx>]

Hi Shawn,

On Wed, 2023-11-29 at 13:18 -0600, Shawn Anastasio wrote:
> On 11/29/23 6:49 AM, Oleksii wrote:
> > On Tue, 2023-11-28 at 15:28 -0600, Shawn Anastasio wrote:
> > > Hi Oleksii,
> > > 
> > > On 11/27/23 8:13 AM, Oleksii Kurochko wrote:
> > > > diff --git a/xen/arch/ppc/include/asm/irq.h
> > > > b/xen/arch/ppc/include/asm/irq.h
> > > > index 5c37d0cf25..49193fddff 100644
> > > > --- a/xen/arch/ppc/include/asm/irq.h
> > > > +++ b/xen/arch/ppc/include/asm/irq.h
> > > > @@ -3,7 +3,9 @@
> > > >  #define __ASM_PPC_IRQ_H__
> > > >  
> > > >  #include <xen/lib.h>
> > > > +#ifdef CONFIG_HAS_DEVICE_TREE
> > > 
> > > I realize that you were likely following PPC's device.h which
> > > also
> > > checks CONFIG_HAS_DEVICE_TREE, but I'm not sure that it makes
> > > sense
> > > to
> > > check this conditional in PPC code at all -- we will always have
> > > HAS_DEVICE_TREE (selected by PPC) and I can't imagine a scenario
> > > where
> > > this will ever not be the case.
> > What about case if ACPI is used? Does ACPI is supported by PPC?
> > 
> > But if you are sure that CONFIG_HAS_DEVICE_TREE will be always
> > selected
> > then I am OK to remove this change.
> > 
> 
> ACPI isn't supported by any PPC platform, we always use device tree.
Thanks for clarification.

Then we can remove useless #ifdef. I'll do it it in next patch version.

> 
> > > 
> > > Unless Jan (or someone else) disagrees, I'd rather this
> > > conditional
> > > be
> > > dropped inside of PPC code.
> > I'll double check but I think I had a compilation issue if it isn't
> > defined.
> > 
> 
> I'm not encountering any issues locally with the conditional dropped,
> but if you are able to reproduce them then let me know.
Sure. I am going to do it tomorrow or today evening if I'll face an
issue will write you.

~ Oleksii