Re: [PATCH -next 07/12] serial: earlycon: Common log banner for command line and DT

[Peter Hurley <peter@hurleysoftware.com>]

Hi Geert,

On 03/05/2015 07:02 AM, Geert Uytterhoeven wrote:
> Hi Peter,
> 
> On Wed, Mar 4, 2015 at 6:24 PM, Peter Hurley <peter@hurleysoftware.com> wrote:
>> --- a/drivers/tty/serial/earlycon.c
>> +++ b/drivers/tty/serial/earlycon.c
>> @@ -69,6 +69,7 @@ static void __init earlycon_init(struct earlycon_device *device,
>>                                  const char *name)
>>  {
>>         struct console *earlycon = device->con;
>> +       struct uart_port *port = &device->port;
>>         const char *s;
>>         size_t len;
>>
>> @@ -82,6 +83,16 @@ static void __init earlycon_init(struct earlycon_device *device,
>>         len = s - name;
>>         strlcpy(earlycon->name, name, min(len + 1, sizeof(earlycon->name)));
>>         earlycon->data = &early_console_dev;
>> +
>> +       if (port->iotype == UPIO_MEM || port->iotype == UPIO_MEM32)
>> +               pr_info("Early serial console at MMIO%s 0x%llx (options '%s')\n",
>> +                       (port->iotype == UPIO_MEM32) ? "32" : "",
>> +                       (unsigned long long)port->mapbase,
> 
> I know you're just moving this piece of code around, but mapbase is a
> resource_size_t (it should actually be phys_addr_t, which probably didn't
> exist in 2007), so it should be printed using %pa and passing a reference,
> instead of %llx and using a cast.

Good point.
I could add that fix to the 8/12 patch which adds the earlycon name to the banner.
I have to respin anyway because of a rebase error that breaks x86 build.

Regards,
Peter Hurley