[PATCH v4] i2c: Add device tree support to i2c-pnx.c

[stigge@antcom.de (Roland Stigge)]

Hi!

On 19/04/12 22:39, Wolfram Sang wrote:
>> If we have a solution soon, I will prepare a new version of the 
>> patch, of course, in the next days.
> 
> Thanks. One question, though: Will it really block dt-conversion? 
> The whole conversion should not be depending on the i2c-driver?

Technically, it would certainly be possible to convert some drivers
and leave others. But I would highly prefer to convert i2c also. With
LPC32xx, there are typically many I2C clients depending on DT on a
reasonable DT enabled mach.

I'm fine with "timeout" also, when you accept it into pnx-i2c.c and
will certainly provide a "timeout" patch update so you can pick the
version which suits best.

Another solution would be to not use timeout with the dt enabled
i2c-pnx for now (using the hard coded default timeout as the current
i2c-pnx.c does) and possibly introduce the (anyway optional) "timeout"
later.

Actually, that would be my favour.

What do you think?

>>> Did you change this, too? Timeouts are better readable in dec
>>> :)
>> 
>> Right. But even when removing the "0x" in the timeout line
>> above, it's still hex, see 
>> Documentation/devicetree/booting-without-of.txt
>> 
>> Or did I get sth. wrong?
> 
> I think the document is probably outdated :( "clock-frequency" is 
> also without 0x and dec.

Interesting! When the documentation is outdated - how does the parser
actually decide between hex (e.g. regs/addresses) and dec (e.g.
clock-frequency) in the absence of "0x"?

Thanks in advance,

Roland