Re: [GIT PULL 1/2] asm-generic: rework PCI I/O space access

[John Garry]

On 17/12/2021 14:32, Arnd Bergmann wrote:
> > I was interesting in the IOPORT_NATIVE flag (or whatever we call it) as
> > it solves the problem of drivers which "unconditionally do inb()/outb()
> > without checking the validity of the address using firmware or other
> > methods first" being built for (and loaded on and crashing) unsuitable
> > systems. Such a problem is in [0]
> >
> > So if we want to support that later, then it seems that someone would
> > need to go back and re-edit many same driver Kconfigs – like hwmon, for
> > example. I think it's better to avoid that and do it now.
> >
> > Arnd, any opinion on that?
> >
> > I'm happy to help with that effort.
> I looked at the options the other day and couldn't really find any that
> fell into this category, so I suggested that Niklas would skip that for the
> moment. 

From looking at the patch Niklas directed us at, as I understand, 
HAS_IOPORT is to decide whether the arch/platform may support PIO 
accessors - inb et al. And on that basis I am confused why it is not 
selected for arm64. And further compounded by:

 config INDIRECT_PIO
	 	bool "Access I/O in non-MMIO mode"
	 	depends on ARM64
	+	depends on HAS_IOPORT

If arm64 does not select, then why depend on it?

> If you have a better way of finding the affected drivers,
> that would be great.

Assuming arm64 should select HAS_IOPORT, I am talking about f71805f as 
an example. According to that patch, this driver additionally depends on 
HAS_IOPORT; however I would rather arm64, like powerpc, should not allow 
that driver to be built at all.

Thanks,
John