[PATCH 2/2] ARM64: kernel: unify ACPI and DT cpus initialization

[Lorenzo Pieralisi]

On Mon, May 11, 2015 at 05:29:45PM +0100, Mark Rutland wrote:
> Hi Lorenzo,
> 
> > +               enable_method = acpi_get_enable_method(cpu);
> > +               if (!enable_method)
> > +                       pr_err("Undefined ACPI enable-method\n");
> 
> I was going to complain that "enable-method" is a DT term that doesn't
> strictly apply to ACPI, but ACPI doesn't seem to have its own
> terminology for this, so I guess this is ok.
> 
> However, I'd recommend replacing "Undefined" with "Unknown/Unsupported"
> in the log message.

Well yes (and that's a problem for cpu_ops header comments too - see other
patch in this series), in ACPI-speak enable-method means nothing, I
think we will have to live with that.

> 
> [...]
> 
> > +static u64 __init of_get_cpu_mpidr(struct device_node *dn)
> > +{
> > +       const u32 *cell;
> > +       u64 hwid;
> > +
> > +       /*
> > +        * A cpu node with missing "reg" property is
> > +        * considered invalid to build a cpu_logical_map
> > +        * entry.
> > +        */
> > +       cell = of_get_property(dn, "reg", NULL);
> > +       if (!cell) {
> > +               pr_err("%s: missing reg property\n", dn->full_name);
> > +               return INVALID_HWID;
> > +       }
> > +
> > +       hwid = of_read_number(cell, of_n_addr_cells(dn));
> 
> Existing bug, but cell should be a __be32 pointer rather than a u32
> pointer (as that's what of_read_number expects).

Will fix it, thanks.

> Other than that, this looks fine to me:
> 
> Acked-by: Mark Rutland <mark.rutland at arm.com>
> Tested-by: Mark Rutland <mark.rutland at arm.com> [DT]

Thank you,
Lorenzo