Re: [RFCv4 05/11] Documentation: DT: omap-ssi binding documentation

From: Sebastian Reichel <sre@debian.org>
To: Tony Lindgren <tony@atomide.com>
Cc: "Linus Walleij" <linus.walleij@linaro.org>,
	"Shubhrajyoti Datta" <omaplinuxkernel@gmail.com>,
	"Carlos Chinea" <cch.devel@gmail.com>,
	"Grant Likely" <grant.likely@linaro.org>,
	"Rob Herring" <rob.herring@calxeda.com>,
	"Pawel Moll" <pawel.moll@arm.com>,
	"Mark Rutland" <mark.rutland@arm.com>,
	"Stephen Warren" <swarren@wwwdotorg.org>,
	"Ian Campbell" <ijc+devicetree@hellion.org.uk>,
	"Rob Landley" <rob@landley.net>,
	"'Benoît Cousson'" <bcousson@baylibre.com>,
	devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-omap@vger.kernel.org, "Pali Rohár" <pali.rohar@gmail.com>,
	"Ивайло Димитров" <freemangordon@abv.bg>,
	"Joni Lapilainen" <joni.lapilainen@gmail.com>,
	"Aaro Koskinen" <aaro.koskinen@iki.fi>
Subject: Re: [RFCv4 05/11] Documentation: DT: omap-ssi binding documentation
Date: Sat, 11 Jan 2014 00:52:09 +0100	[thread overview]
Message-ID: <20140110235208.GA16048@earth.universe> (raw)
In-Reply-To: <20131219190344.GW27438@atomide.com>

[-- Attachment #1: Type: text/plain, Size: 2131 bytes --]

Hi Tony,

On Thu, Dec 19, 2013 at 11:03:44AM -0800, Tony Lindgren wrote:
> > +Required properties:
> > +- compatible:		Should include "ti,omap3-ssi".
> > +- reg-names:		Contains the values "sys" and "gdd".
> 
> Do you need the reg-names? The order won't change so you can just
> document the order in the binding?

The names are not needed, but I like self-documenting code/bindings.

Also the examples in Documentation/devicetree/bindings/resource-names.txt
look similar to this case.

What do you have against the -names properties?

The same statement goes for the following -names comments, so I will
skip them ;)

> [...]
> > +- ti,ssi-cawake-gpio:	Defines which GPIO pin is used to signify CAWAKE
> > +			events for the port. This is an optional board-specific
> > +			property. If it's missing the port will not be
> > +			enabled.
> 
> Hmm this might be just a wake-up GPIO? If so, you should be able to
> just set it up as an interrupt and do a request_irq on the pinctrl-single
> entry for it.

Yes, this gpio is used as interrupt in the driver, but its also read
directly. I already considered making it an irq in the DT data
(since its mainly used as irq), but I could not find out how to read
the current status of an irq line.

> It might even be one of the already mapped interrupt lines that the code is
> remuxing to a GPIO for idle? If so, then you can just use the new binding
> for interrupts-extended to handle the wake-up events.
> 
> If you post the GPIO number for ti,ssi-cawake-gpio and the interrupt
> numbers I can check if there's a need to handle it separately as a GPIO
> pin or if it already can be automatically handled for the wake-up events.

You can see it in one of the next patches, which adds the needed
nodes in omap3-n900.dts. The used GPIO on N900 is 151 (gpio5 23)
and I use the following pinmux configuration:

0x152 (PIN_INPUT | WAKEUP_EN | MUX_MODE4)

P.S.: I intend to get this into 3.15. Before I will send an updated
      series, which uses the omap clock DT bindings as requested by
      DT binding maintainers.

-- Sebastian

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]