All of lore.kernel.org
 help / color / mirror / Atom feed
From: Tomasz Figa <t.figa@samsung.com>
To: devicetree-discuss@lists.ozlabs.org
Cc: Tushar Behera <tushar.behera@linaro.org>,
	Girish KS <girishks2000@gmail.com>,
	"linux-samsung-soc@vger.kernel.org"
	<linux-samsung-soc@vger.kernel.org>,
	"patches@linaro.org" <patches@linaro.org>,
	dtor@mail.ru, Manish Badarkhe <badarkhe.manish@gmail.com>,
	"kgene.kim@samsung.com" <kgene.kim@samsung.com>,
	david@protonic.nl, "Kumar, Anil" <anilkumar.v@ti.com>,
	"linux-arm-kernel@lists.infradead.org"
	<linux-arm-kernel@lists.infradead.org>
Subject: Re: [PATCH 1/5] ARM: dts: exynos5250-arndale: Add node entry for gpio-buttons
Date: Thu, 07 Feb 2013 12:25:25 +0100	[thread overview]
Message-ID: <18141670.7lKVRsjORY@amdc1227> (raw)
In-Reply-To: <5113768B.90102@linaro.org>

On Thursday 07 of February 2013 15:10:27 Tushar Behera wrote:
> +CC: David Jander <david@protonic.nl>
> +CC: Dmitry Torokhov <dtor@mail.ru>
> 
> On 02/07/2013 01:13 PM, Girish KS wrote:
> >>>>>> +
> >>>>>> +    gpio_keys {
> >>>>>> +            compatible = "gpio-keys";
> >>>>>> +            #address-cells = <1>;
> >>>>>> +            #size-cells = <0>;
> >>>>> 
> >>>>> Just want to understand why these properties are here?
> >>>>> As these properties are for child dt node. But have not seen
> >>>>> anyone is using here.
> >>>> 
> >>>> That is how gpio_keys node entries are defined in other .dts files.
> >>> 
> >>> I have gone through example for address-cells and size-cells in
> >>> following link:
> >>> http://devicetree.org/mediawiki/index.php?title=Device_Tree_Usage&s
> >>> table=1#CPU_addressing
> >>> 
> >>> which indicates that these fields are for child "reg".
> >>> I think, here in child node there is no "reg". so there is no use
> >>> of address-cells and size-cells propeties.
> >> 
> >> Please check Documentation/devicetree/bindings/gpio/gpio_keys.txt
> >> 
> >> And whether these properties are required or not, I will let device
> >> tree experts to comment on that.
> >> 
> >> As such, currently all node entries for gpio_keys use these
> >> properties.
> > 
> > you can just verify by a simple test.
> > delete the 2 lines. address-cells and size cells.
> > this will have no affect.
> > now with these 2 lines deleted lines add a dummy reg property with
> > address and cell. dtb compiler will warn.
> > So I think what manish reffered is right.
> > There is also a reference of smdk4210 for kepads.
> 
> Right. Compilation is ok even after removing the address-cell and
> size-cell properties.
> 
> But since this is used across all the instances, I would like to know
> the view of the authors on this.
> 
> David, Dimitry,
> 
> Can you please let us know your opinion on this? If these properties are
> not required, then we can remove them from the documentation and from
> node entries in several other dts files.

The #address-cells and #size-cells are used only together with reg 
properties of sub nodes. They define how many cells in reg specifier are 
used for registers address and length.

So, if you have #address-cells = <1> and #size-cells = <0>, then your reg 
specifier will be simply <addr>. You can get more complex specifiers using 
more address cells and size cells, like:

#address-cells = <3>;
#size-cells = <1>;

subdev@1,2,3 {
	reg = <1 2 3 0x1000>;
	/* ... */
};

If you don't intend to use reg property for child device addressing then 
you don't define #address-cells and #size-cells properties.

I think that the documentation of gpio_keys binding should be corrected 
and possibly also moved to bindings/input instead of bindings/gpio, as 
this directory is intended to be used for GPIO controllers, not GPIO 
consumers.

Best regards,
-- 
Tomasz Figa
Samsung Poland R&D Center
SW Solution Development, Linux Platform

WARNING: multiple messages have this Message-ID (diff)
From: t.figa@samsung.com (Tomasz Figa)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 1/5] ARM: dts: exynos5250-arndale: Add node entry for gpio-buttons
Date: Thu, 07 Feb 2013 12:25:25 +0100	[thread overview]
Message-ID: <18141670.7lKVRsjORY@amdc1227> (raw)
In-Reply-To: <5113768B.90102@linaro.org>

On Thursday 07 of February 2013 15:10:27 Tushar Behera wrote:
> +CC: David Jander <david@protonic.nl>
> +CC: Dmitry Torokhov <dtor@mail.ru>
> 
> On 02/07/2013 01:13 PM, Girish KS wrote:
> >>>>>> +
> >>>>>> +    gpio_keys {
> >>>>>> +            compatible = "gpio-keys";
> >>>>>> +            #address-cells = <1>;
> >>>>>> +            #size-cells = <0>;
> >>>>> 
> >>>>> Just want to understand why these properties are here?
> >>>>> As these properties are for child dt node. But have not seen
> >>>>> anyone is using here.
> >>>> 
> >>>> That is how gpio_keys node entries are defined in other .dts files.
> >>> 
> >>> I have gone through example for address-cells and size-cells in
> >>> following link:
> >>> http://devicetree.org/mediawiki/index.php?title=Device_Tree_Usage&s
> >>> table=1#CPU_addressing
> >>> 
> >>> which indicates that these fields are for child "reg".
> >>> I think, here in child node there is no "reg". so there is no use
> >>> of address-cells and size-cells propeties.
> >> 
> >> Please check Documentation/devicetree/bindings/gpio/gpio_keys.txt
> >> 
> >> And whether these properties are required or not, I will let device
> >> tree experts to comment on that.
> >> 
> >> As such, currently all node entries for gpio_keys use these
> >> properties.
> > 
> > you can just verify by a simple test.
> > delete the 2 lines. address-cells and size cells.
> > this will have no affect.
> > now with these 2 lines deleted lines add a dummy reg property with
> > address and cell. dtb compiler will warn.
> > So I think what manish reffered is right.
> > There is also a reference of smdk4210 for kepads.
> 
> Right. Compilation is ok even after removing the address-cell and
> size-cell properties.
> 
> But since this is used across all the instances, I would like to know
> the view of the authors on this.
> 
> David, Dimitry,
> 
> Can you please let us know your opinion on this? If these properties are
> not required, then we can remove them from the documentation and from
> node entries in several other dts files.

The #address-cells and #size-cells are used only together with reg 
properties of sub nodes. They define how many cells in reg specifier are 
used for registers address and length.

So, if you have #address-cells = <1> and #size-cells = <0>, then your reg 
specifier will be simply <addr>. You can get more complex specifiers using 
more address cells and size cells, like:

#address-cells = <3>;
#size-cells = <1>;

subdev at 1,2,3 {
	reg = <1 2 3 0x1000>;
	/* ... */
};

If you don't intend to use reg property for child device addressing then 
you don't define #address-cells and #size-cells properties.

I think that the documentation of gpio_keys binding should be corrected 
and possibly also moved to bindings/input instead of bindings/gpio, as 
this directory is intended to be used for GPIO controllers, not GPIO 
consumers.

Best regards,
-- 
Tomasz Figa
Samsung Poland R&D Center
SW Solution Development, Linux Platform

  reply	other threads:[~2013-02-07 11:25 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-02-07  5:15 [PATCH 1/5] ARM: dts: exynos5250-arndale: Add node entry for gpio-buttons Tushar Behera
2013-02-07  5:15 ` Tushar Behera
2013-02-07  5:15 ` [PATCH 2/5] ARM: dts: exynos5250-arndale: Add PMIC node entry Tushar Behera
2013-02-07  5:15   ` Tushar Behera
2013-02-07  5:49   ` Kumar, Anil
2013-02-07  5:49     ` Kumar, Anil
2013-02-07  5:56     ` Tushar Behera
2013-02-07  5:56       ` Tushar Behera
2013-02-11  0:38       ` amit daniel kachhap
2013-02-11  0:38         ` amit daniel kachhap
2013-03-11  2:05         ` Kukjin Kim
2013-03-11  2:05           ` Kukjin Kim
2013-02-07  5:15 ` [PATCH 3/5] ARM: dts: exynos5250-arndale: Add vmmc regulator support Tushar Behera
2013-02-07  5:15   ` Tushar Behera
2013-02-07  5:15 ` [PATCH 4/5] ARM: dts: exynos5250-arndale: Add MFC codec support Tushar Behera
2013-02-07  5:15   ` Tushar Behera
2013-02-07  5:15 ` [PATCH 5/5] ARM: dts: exynos5250-arndale: Add HDMI HPD and regulator node Tushar Behera
2013-02-07  5:15   ` Tushar Behera
2013-02-07  6:17 ` [PATCH 1/5] ARM: dts: exynos5250-arndale: Add node entry for gpio-buttons Kumar, Anil
2013-02-07  6:17   ` Kumar, Anil
2013-02-07  6:26   ` Tushar Behera
2013-02-07  6:26     ` Tushar Behera
2013-02-07  6:36     ` Manish Badarkhe
2013-02-07  6:36       ` Manish Badarkhe
2013-02-07  6:56       ` Tushar Behera
2013-02-07  6:56         ` Tushar Behera
2013-02-07  7:43         ` Girish KS
2013-02-07  7:43           ` Girish KS
2013-02-07  9:40           ` Tushar Behera
2013-02-07  9:40             ` Tushar Behera
2013-02-07 11:25             ` Tomasz Figa [this message]
2013-02-07 11:25               ` Tomasz Figa
2013-02-07 10:30 ` Tomasz Figa
2013-02-07 10:30   ` Tomasz Figa
2013-03-11  2:05   ` Kukjin Kim
2013-03-11  2:05     ` Kukjin Kim
2013-03-11  9:44     ` Tushar Behera
2013-03-11  9:44       ` Tushar Behera

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=18141670.7lKVRsjORY@amdc1227 \
    --to=t.figa@samsung.com \
    --cc=anilkumar.v@ti.com \
    --cc=badarkhe.manish@gmail.com \
    --cc=david@protonic.nl \
    --cc=devicetree-discuss@lists.ozlabs.org \
    --cc=dtor@mail.ru \
    --cc=girishks2000@gmail.com \
    --cc=kgene.kim@samsung.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-samsung-soc@vger.kernel.org \
    --cc=patches@linaro.org \
    --cc=tushar.behera@linaro.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.