[3/6] irqchip: irq-st: Add documentation for STi based syscfg IRQs

Signed-off-by: Lee Jones <lee.jones@linaro.org>
---
 .../interrupt-controller/st,sti-irq-syscfg.txt     | 35 ++++++++++++++++++++++
 1 file changed, 35 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/interrupt-controller/st,sti-irq-syscfg.txt

Hey Lee,

Rob, Grant, question for you below:

On Tue, Jul 29, 2014 at 03:05:41PM +0100, Lee Jones wrote:
> Signed-off-by: Lee Jones <lee.jones@linaro.org>
> ---
>  .../interrupt-controller/st,sti-irq-syscfg.txt     | 35 ++++++++++++++++++++++
>  1 file changed, 35 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/interrupt-controller/st,sti-irq-syscfg.txt
> 
> diff --git a/Documentation/devicetree/bindings/interrupt-controller/st,sti-irq-syscfg.txt b/Documentation/devicetree/bindings/interrupt-controller/st,sti-irq-syscfg.txt
> new file mode 100644
> index 0000000..ced6014
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/interrupt-controller/st,sti-irq-syscfg.txt
> @@ -0,0 +1,35 @@
> +STMicroelectronics STi System Configuration Controlled IRQs
> +-----------------------------------------------------------
> +
> +On STi based systems; External, CTI (Core Sight), PMU (Performance Management),
> +and PL310 L2 Cache IRQs are controlled using System Configuration registers.
> +This driver is used to unmask them prior to use.

We don't usually refer to the driver in binding docs as they are
supposed to be OS-agnostic.

> +
> +Required properties:
> +- compatible	: Should be set to one of:
> +			"st,stih415-irq-syscfg"
> +			"st,stih416-irq-syscfg"
> +			"st,stih407-irq-syscfg"
> +			"st,stid127-irq-syscfg"
> +- st,syscfg	: Phandle to Cortex-A9 IRQ system config registers
> +- st,irq-device	: Array of IRQs to enable - should be 2 in length
> +- st,fiq-device	: Array of FIQs to enable - should be 2 in length
> +
> +Optional properties:
> +- st,invert-ext	: External IRQs can be inverted at will.  This property inverts
> +		  these IRQs using bitwise logic.  A number of defines have been
> +		  provided for convenience:
> +			ST_IRQ_SYSCFG_EXT_1_INV
> +			ST_IRQ_SYSCFG_EXT_2_INV
> +			ST_IRQ_SYSCFG_EXT_3_INV
> +Example:

Rob, Grant, should we be mingling implementation details (macros from
include files) with the binding documentation?  It just _seems_ wrong
too me.  But if we agree that the binding docs, the dts files, and their
includes are always kept together, I also don't see the harm...

> +
> +irq-syscfg {
> +	compatible    = "st,stih416-irq-syscfg";
> +	st,syscfg     = <&syscfg_cpu>;
> +	st,irq-device = <ST_IRQ_SYSCFG_PMU_0>,
> +			<ST_IRQ_SYSCFG_PMU_1>;
> +	st,fiq-device = <ST_IRQ_SYSCFG_DISABLED>,
> +			<ST_IRQ_SYSCFG_DISABLED>;
> +	st,invert-ext = <(ST_IRQ_SYSCFG_EXT_1_INV | ST_IRQ_SYSCFG_EXT_3_INV)>;
> +};

thx,

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

On Mon, 18 Aug 2014, Jason Cooper wrote:

> Hey Lee,
> 
> Rob, Grant, question for you below:
> 
> On Tue, Jul 29, 2014 at 03:05:41PM +0100, Lee Jones wrote:
> > Signed-off-by: Lee Jones <lee.jones@linaro.org>
> > ---
> >  .../interrupt-controller/st,sti-irq-syscfg.txt     | 35 ++++++++++++++++++++++
> >  1 file changed, 35 insertions(+)
> >  create mode 100644 Documentation/devicetree/bindings/interrupt-controller/st,sti-irq-syscfg.txt
> > 
> > diff --git a/Documentation/devicetree/bindings/interrupt-controller/st,sti-irq-syscfg.txt b/Documentation/devicetree/bindings/interrupt-controller/st,sti-irq-syscfg.txt
> > new file mode 100644
> > index 0000000..ced6014
> > --- /dev/null
> > +++ b/Documentation/devicetree/bindings/interrupt-controller/st,sti-irq-syscfg.txt
> > @@ -0,0 +1,35 @@
> > +STMicroelectronics STi System Configuration Controlled IRQs
> > +-----------------------------------------------------------
> > +
> > +On STi based systems; External, CTI (Core Sight), PMU (Performance Management),
> > +and PL310 L2 Cache IRQs are controlled using System Configuration registers.
> > +This driver is used to unmask them prior to use.
> 
> We don't usually refer to the driver in binding docs as they are
> supposed to be OS-agnostic.

I beg to differ:

$ git grep -i driver -- Documentation/devicetree/ | wc -l
355

I can understand if I was referencing Linux driver implementation
specifics, but any of the OS's drivers would also be un/masking IRQs
in order to provide this functionality.


> > +Required properties:
> > +- compatible	: Should be set to one of:
> > +			"st,stih415-irq-syscfg"
> > +			"st,stih416-irq-syscfg"
> > +			"st,stih407-irq-syscfg"
> > +			"st,stid127-irq-syscfg"
> > +- st,syscfg	: Phandle to Cortex-A9 IRQ system config registers
> > +- st,irq-device	: Array of IRQs to enable - should be 2 in length
> > +- st,fiq-device	: Array of FIQs to enable - should be 2 in length
> > +
> > +Optional properties:
> > +- st,invert-ext	: External IRQs can be inverted at will.  This property inverts
> > +		  these IRQs using bitwise logic.  A number of defines have been
> > +		  provided for convenience:
> > +			ST_IRQ_SYSCFG_EXT_1_INV
> > +			ST_IRQ_SYSCFG_EXT_2_INV
> > +			ST_IRQ_SYSCFG_EXT_3_INV
> > +Example:
> 
> Rob, Grant, should we be mingling implementation details (macros from
> include files) with the binding documentation?  It just _seems_ wrong
> too me.  But if we agree that the binding docs, the dts files, and their
> includes are always kept together, I also don't see the harm...

IMHO, we should be able to document anything here which appears in the
DTS(I) files.

> > +irq-syscfg {
> > +	compatible    = "st,stih416-irq-syscfg";
> > +	st,syscfg     = <&syscfg_cpu>;
> > +	st,irq-device = <ST_IRQ_SYSCFG_PMU_0>,
> > +			<ST_IRQ_SYSCFG_PMU_1>;
> > +	st,fiq-device = <ST_IRQ_SYSCFG_DISABLED>,
> > +			<ST_IRQ_SYSCFG_DISABLED>;
> > +	st,invert-ext = <(ST_IRQ_SYSCFG_EXT_1_INV | ST_IRQ_SYSCFG_EXT_3_INV)>;
> > +};