[PATCH 5/5] arm64: add early_ioremap support

[robherring2@gmail.com (Rob Herring)]

On Wed, Mar 5, 2014 at 10:38 AM, Mark Salter <msalter@redhat.com> wrote:
> On Tue, 2014-03-04 at 17:31 -0600, Rob Herring wrote:
>> On Tue, Mar 4, 2014 at 2:08 PM, Mark Salter <msalter@redhat.com> wrote:
>> > Add support for early IO or memory mappings which are needed
>> > before the normal ioremap() is usable. This also adds fixmap
>> > support for permanent fixed mappings such as that used by the
>> > earlyprintk device register region.
>>
>> One minor comment:
>>
>> > +enum fixed_addresses {
>> > +       FIX_EARLYCON,
>>
>> Can you align this name with x86 and rename to FIX_EARLYCON_MEM_BASE.
>> Doing that will help enable the earlycon driver in 8250_early.c.
>
> Is there any advantage to earlycon=uart8250 vs. earlyprintk=uart8250?

Functionally, they are pretty much the same. I'm working on a series
to generalize earlycon and then we will be able to delete
early_printk.c from arm64 and ultimately share with arm. Doing this
alignment just saves a patch to change it later.

> I don't mind using FIX_EARLYCON_MEM_BASE, but CONFIG_FIX_EARLYCON_MEM
> is X86 only right now and I think changing that should be a separate
> patch.

Yes, definitely.

Rob