Re: [PATCH] of: Custom printk format specifier for device node

[Joe Perches <joe@perches.com>]

On Tue, 2015-01-20 at 20:06 +0200, Pantelis Antoniou wrote:
> On Jan 20, 2015, at 19:59 , Joe Perches <joe@perches.com> wrote:
> > On Tue, 2015-01-20 at 16:52 +0200, Pantelis Antoniou wrote:
> >>> On Jan 20, 2015, at 16:47 , Rob Herring <robherring2@gmail.com> wrote:
> >>> On Tue, Jan 20, 2015 at 8:34 AM, Pantelis Antoniou
> >>> <pantelis.antoniou@konsulko.com> wrote:
> >>>> 90% of the usage of device node's full_name is printing it out
> >>>> in a kernel message. Preparing for the eventual delayed allocation
> >>>> introduce a custom printk format specifier that is both more
> >>>> compact and more pleasant to the eye.
> > []
> >>>> diff --git a/Documentation/printk-formats.txt b/Documentation/printk-formats.txt
> > []
> >>>> +Device tree nodes:
> >>>> +
> >>>> +       %pO{,1,2}
> >>> 
> >>> 'O' is not very obvious, but I imagine we are somewhat limted in our
> >>> choice here?
> >>> 
> >> 
> >> All the good women are married, all the handsome men are gay, all the obvious
> >> format specifiers are taken.
> > 
> > Not really at all.
> > 
> > I quite dislike '0' as the format type specifier for
> > a device tree node as there's no mnemonic mapping.
> > 
> There’s only so many characters one can use; [DdNn] are taken. And it’s O != 0.

The '[Dd]' entry type doesn't just have to be used for dentries.
It could also be used for Device Tree Nodes via 'DTN[x]'

'T' for tree could also be used.