Re: [PATCH net-next 14/15] net: dsa: sja1105: replace mdiobus-pcs with xpcs-plat driver

[Vladimir Oltean]

On Wed, Nov 19, 2025 at 12:05:43PM +0000, Russell King (Oracle) wrote:
> On Wed, Nov 19, 2025 at 12:03:29PM +0000, Russell King (Oracle) wrote:
> > On Wed, Nov 19, 2025 at 02:01:11PM +0200, Vladimir Oltean wrote:
> > > I do wonder how to print resource_size_t (typedef to phys_addr_t, which
> > > is typedeffed to u64 or u32 depending on CONFIG_PHYS_ADDR_T_64BIT).
> > 
> > From the now hard to find Documentation/core-api/printk-formats.rst:

I really wasn't aware of this, and I was reading the inline docs from
lib/vsprintf.c all this time... Thanks!

> > Physical address types phys_addr_t
> > ----------------------------------
> > 
> > ::
> > 
> >         %pa[p]  0x01234567 or 0x0123456789abcdef
> > 
> > For printing a phys_addr_t type (and its derivatives, such as
> > resource_size_t) which can vary based on build options, regardless of the
> > width of the CPU data path.
> > 
> > Passed by reference.
> 
> Hmm, but I guess you don't want the 0x prefix. Maybe print it to a
> separate buffer and then lop off the first two characters?
> 
> Another solution would be to always cast it to a u64 and use %llx as
> suggested in the "Integer types" section.

Yeah, although I see the ePAPR section on node names doesn't disallow
the 0x prefix for the unit-address, it is just that the established
convention is without it.

By far the most unassuming option seems to be to do explicit type
casting to unsigned long long. I've started a new test build with the
changes made so far.