Message-ID: <5665F5DF.50002@collabora.co.uk>
Date: Mon, 07 Dec 2015 21:10:55 +0000
From: Martyn Welch <martyn.welch@collabora.co.uk>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Icedove/31.8.0
MIME-Version: 1.0
To: Rob Herring <robh@kernel.org>, Linus Walleij <linus.walleij@linaro.org>
CC: Arnd Bergmann <arnd@arndb.de>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Pawel Moll <pawel.moll@arm.com>, Mark Rutland <mark.rutland@arm.com>,
        Ian Campbell <ijc+devicetree@hellion.org.uk>,
        Kumar Gala <galak@codeaurora.org>,
        Russell King <linux@arm.linux.org.uk>, Kukjin Kim <kgene@kernel.org>,
        Krzysztof Kozlowski <k.kozlowski@samsung.com>,
        devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
        linux-arm-kernel@lists.infradead.org,
        linux-samsung-soc@vger.kernel.org, Olof Johansson <olof@lixom.net>
Subject: Re: [PATCH 1/3] Device tree binding documentation for gpio-switch
References: <1449250275-23435-1-git-send-email-martyn.welch@collabora.co.uk> <1449250275-23435-2-git-send-email-martyn.welch@collabora.co.uk> <20151207173702.GA17659@rob-hp-laptop>
In-Reply-To: <20151207173702.GA17659@rob-hp-laptop>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 3988
Lines: 109


On 07/12/15 17:37, Rob Herring wrote:
> +Linus W
>
> On Fri, Dec 04, 2015 at 05:31:13PM +0000, Martyn Welch wrote:
>> This patch adds documentation for the gpio-switch binding. This binding
>> provides a mechanism to bind named links to gpio, with the primary
>> purpose of enabling standardised access to switches that might be standard
>> across a group of devices but implemented differently on each device.
>
> This is good and what I suggested, but it now makes me wonder if switch
> is generic enough. This boils down to needing to expose single gpio
> lines to userspace with a defined function/use. IIRC, there's been some
> discussion about this before along with improving the userspace
> interface for GPIO in general. So I'd like to get Linus' thoughts on
> this.
>

No problem. Rename gpio-signal?

>
>> Signed-off-by: Martyn Welch <martyn.welch@collabora.co.uk>
>> ---
>>   .../devicetree/bindings/misc/gpio-switch.txt       | 47 ++++++++++++++++++++++
>>   1 file changed, 47 insertions(+)
>>   create mode 100644 Documentation/devicetree/bindings/misc/gpio-switch.txt
>>
>> diff --git a/Documentation/devicetree/bindings/misc/gpio-switch.txt b/Documentation/devicetree/bindings/misc/gpio-switch.txt
>> new file mode 100644
>> index 0000000..13528bd
>> --- /dev/null
>> +++ b/Documentation/devicetree/bindings/misc/gpio-switch.txt
>> @@ -0,0 +1,47 @@
>> +Device-Tree bindings for gpio attached switches.
>> +
>> +This provides a mechanism to provide a named link to specified gpios. This can
>> +be useful in instances such as when theres a need to monitor a switch, which is
>> +common across a family of devices, but attached to different gpios and even
>> +implemented in different ways on differnet devices.
>> +
>> +Required properties:
>> +	- compatible = "gpio-switch";
>> +
>> +Each signal is represented as a sub-node of "gpio-switch". The naming of
>> +sub-nodes is arbitrary.
>> +
>> +Required sub-node properties:
>> +
>> +	- label: Name to be given to gpio switch.
>> +	- gpios: OF device-tree gpio specification.
>> +
>> +Optional sub-node properties:
>> +
>> +	- read-only: Boolean flag to mark the gpio as read-only, i.e. the line
>> +	  should not be driven by the host.
>
> In terms a a switch use, allowing driving it would be an override of the
> switch. Is that the idea here?
>

Yeah - since it had become a lot more generic and a lot of 
switches/signals would probably be implemented with a pull-up resistor 
of something like that, it seemed to make sense to allow them to be 
driven as well.

>> +
>> +Example nodes:
>> +
>> +        gpio-switch {
>> +                compatible = "gpio-switch";
>
> Both from a binding and driver perspective, there is no point in
> grouping these. Each node can simply have this compatible string.
>

True. I did it this way as this is how gpio-keys is implemented. OK, 
that has one optional parameter (autorepeat) for the block and this has 
none. Though I can also see that these probably have less in common than 
the individual lines used in gpio-keys.

>> +
>> +                write-protect {
>> +                        label = "write-protect";
>> +                        gpios = <&gpx3 0 GPIO_ACTIVE_LOW>;
>> +                        read-only;
>> +                };
>> +
>> +                developer-switch {
>> +                        label = "developer-switch";
>> +                        gpios = <&gpx1 3 GPIO_ACTIVE_HIGH>;
>> +                        read-only;
>> +                };
>> +
>> +                recovery-switch {
>> +                        label = "recovery-switch";
>> +                        gpios = <&gpx0 7 GPIO_ACTIVE_LOW>;
>> +                        read-only;
>> +                };
>> +        };
>> +
>> --
>> 2.1.4
>>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/