Re: [PATCH 1/2] i2c: qup: Add device tree bindings information

[Mark Rutland <mark.rutland-5wv7dgnIgG8@public.gmane.org>]

On Fri, Sep 13, 2013 at 10:13:15AM +0100, Ivan T. Ivanov wrote:
> Hi Mark, 
> 
> On Thu, 2013-09-12 at 17:28 +0100, Mark Rutland wrote: 
> > On Thu, Aug 29, 2013 at 02:27:52PM +0100, Ivan T. Ivanov wrote:
> > > From: "Ivan T. Ivanov" <iivanov-NEYub+7Iv8PQT0dZR+AlfA@public.gmane.org>
> > > 
> > > The Qualcomm Universal Peripherial (QUP) wraps I2C mini-core and
> > > provide input and output FIFO's for it. I2C controller can operate
> > > as master with supported bus speeds of 100Kbps and 400Kbps.
> > > 
> > > Signed-off-by: Ivan T. Ivanov <iivanov-NEYub+7Iv8PQT0dZR+AlfA@public.gmane.org>
> > > ---
> > >  Documentation/devicetree/bindings/i2c/i2c-qup.txt |   99 +++++++++++++++++++++
> > >  1 file changed, 99 insertions(+)
> > >  create mode 100644 Documentation/devicetree/bindings/i2c/i2c-qup.txt
> > > 
> > > diff --git a/Documentation/devicetree/bindings/i2c/i2c-qup.txt b/Documentation/devicetree/bindings/i2c/i2c-qup.txt
> > > new file mode 100644
> > > index 0000000..c682726
> > > --- /dev/null
> > > +++ b/Documentation/devicetree/bindings/i2c/i2c-qup.txt
> > > @@ -0,0 +1,99 @@
> > > +Qualcomm Universal Periferial (QUP) I2C controller

s/Periferial/Peripheral/ ?

> > > +
> > > +Required properties:
> > > + - compatible : should be "qcom,i2c-qup"
> > > + - reg : Offset and length of the register region for the device
> > > + - interrupts : core interrupt
> > 
> > How about the following:
> > 
> > - interrupts: interrupt-specifier for the core interrupt.
> 
> Ok.
> 
> > 
> > > +
> > > + - pinctrl-names: Should contain only one value - "default".
> > > + - pinctrl-0: Should specify pin control group used for this controller.
> > > +
> > > + - clocks : phandles to clock instances of the device tree nodes
> > 
> > Clocks aren't just phandles, they have a clock-specifier component. This
> > should probably be something like:
> > 
> >  - clocks: a list of phandle + clock-specifier pairs for each entry in
> >            clock-names
> 
> Right now MSM clocks [1] are just phandles, I should add #clock-cells = <0>.

I see no reason to strongly tie the binding to a particular set of
clocks when it's easy to make it generic. What if a new platform uses
QUP, but its clocks have a number of clock-cells? As clock-specifiers
may be zero cells, describing this as a phandle + clock-specifier pair
means this will work with any clock provider following the common clock
bindings.

> 
> > 
> > > + - clock-names :
> > > +		"core" : Allow access to FIFO buffers and registers
> > 
> > Huh? That description doesn't seem to descripe the hardware.
> 
> How about: Allow access to I2C core FIFO buffers and registers.

My only concern here is the words "allow access", as that seems odd in
the context of a hardware description.

It seems most bindings don't describe what the clocks do as that's
defined in the manual for a piece of hardware anyway, so unfortunately I
don't have a good example.

I think something like the below would be more in keeping with what's
expected:

		"core": Clock controlling the FIFO buffers and registers

> 
> > 
> > > +		"iface" : Clock used by QUP interface
> > 
> > Which interface? The slave interface the CPUs access, or the interface
> > to the I2C devices?
> 
> 
> My understanding for this is that QUP is wrapping I2C core. QUP is
> interface to I2C core.

Ok.

> 
> > 
> > Are these the only clock inputs to the device?
> 
> Yep.

Ok.

> 
> > 
> > Is there a regulator input that might need to be specified?
> > 
> 
> No. Power management is done via clock gating.

Ok.

> 
> > > +
> > > + - #address-cells : should be <1> Address cells for I2C device address
> > > + - #size-cells : should be <0> I2C addresses have no size component.
> > > +
> > > +Optional properties :
> > > + - Child nodes conforming to i2c bus binding
> > > + - clock-frequency : Desired I2C bus clock frequency in Hz. If
> > > +		not set thedefault frequency is 100kHz
> > 
> > Why is this necessary?
> 
> This is how I2C bus frequency could be specified. It is standard
> I2C property. Not sure that I get the question.

Not all i2c bindings have this, but I see many do. Never mind.

> 
> > 
> > > + - qcom,src-freq : Frequency of the source clocking this bus in Hz.
> > > +		Divider value is set based on soruce-frequency and
> > > +		desired I2C bus frequency. If this value is not
> > > +		provided, the source clock is assumed to be running
> > > +		at 19.2 MHz.
> > 
> > This looks like it should be a clock input.
> 
> Thanks, I will change it.

Great!

Cheers,
Mark.
--
To unsubscribe from this list: send the line "unsubscribe devicetree" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html