Re: [PATCH v5 2/2] xen/pci: replace call to is_memory_hole to pci_check_bar

[Julien Grall <julien@xxxxxxx>]

Hi Stefano,

On 08/09/2022 21:59, Stefano Stabellini wrote:
> > +/*
> > + * TODO: BAR addresses and Root Complex window addresses are not guaranteed
> > + * to be page aligned. We should check for alignment but this is not the
> > + * right place for alignment check.
> > + */
> > +static int is_bar_valid(const struct dt_device_node *dev,
> > +                        uint64_t addr, uint64_t len, void *data)
> > +{
> > +    struct pdev_bar_check *bar_data = data;
> > +    unsigned long s = bar_data->start;
> > +    unsigned long e = bar_data->end;
> > +
> > +    if ( (s >= addr) && (e <= (addr + len - 1)) )
> > +        bar_data->is_valid =  true;
> "s" and "e" are "unsigned long" while "addr" and "len" are uint64_t. Is
> that OK?
Good catch. No, physical address on Arm32 can be up to 40 bits.

> Specifically, considering a potential arm32 case, shouldn't "s" and "e"
> be uint64_t as well? Which means pdev_bar_check.start and end should be
> uint64_t?
They should be paddr_t which will be 64-bit on both arm32 and arm64.

Cheers,

--
Julien Grall