On 03/22/2016 02:15 PM, Rob Herring wrote:

On Tue, Mar 22, 2016 at 5:46 AM, Aleksey Makarov
[off-list ref] wrote:

quoted

From: Leif Lindholm <redacted>

We have multiple "earlycon" early_param handlers - merge the DT one into
the main earlycon one.  It's a cleanup that also will be useful
to decide if ACPI SPCR earlycon should be set up.

How so? Isn't that determined by whether we have ACPI tables or a DT?

Oops, I missed this.  This patch was not only to do parsing in one place, 
but also to allow addition of code that decide which of ACPI/DT should be used
for earlycon initialization based on acpi_disabled.

I will fix this, thank you.

This goes against trying to limit places of FDT parsing and keeping
the early scanning all within fdt.c. At least it is not in the arch
code. That said, I'm okay with this change if it helps.

Rob