Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: Re: [PATCH 1/5] dt-bindings: connector: add optional properties for
 Type-B
To:     Chunfeng Yun <chunfeng.yun@mediatek.com>
Cc:     Rob Herring <robh+dt@kernel.org>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Heikki Krogerus <heikki.krogerus@linux.intel.com>,
        Mark Rutland <mark.rutland@arm.com>,
        Matthias Brugger <matthias.bgg@gmail.com>,
        Adam Thomson <Adam.Thomson.Opensource@diasemi.com>,
        Li Jun <jun.li@nxp.com>,
        Badhri Jagan Sridharan <badhri@google.com>,
        Andy Shevchenko <andy.shevchenko@gmail.com>,
        Min Guo <min.guo@mediatek.com>, devicetree@vger.kernel.org,
        linux-kernel@vger.kernel.org, linux-usb@vger.kernel.org,
        linux-arm-kernel@lists.infradead.org,
        linux-mediatek@lists.infradead.org
References: <1552025622-15582-1-git-send-email-chunfeng.yun@mediatek.com>
 <1552025622-15582-2-git-send-email-chunfeng.yun@mediatek.com>
 <693b11e7-fa9e-ab1f-561a-beb758b1872f@redhat.com>
 <1552282383.10179.21.camel@mhfsdcap03>
 <c8084259-74d3-c85f-1275-854826c935a0@redhat.com>
 <1552360688.10179.55.camel@mhfsdcap03>
 <464f1021-4c5e-9137-489d-65fd578575d3@redhat.com>
 <1552472109.10179.103.camel@mhfsdcap03>
From:   Hans de Goede <hdegoede@redhat.com>
Message-ID: <0c3ca095-9077-0b9d-839f-61ebc066e3f9@redhat.com>
Date:   Wed, 13 Mar 2019 11:18:37 +0100
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101
 Thunderbird/60.4.0
MIME-Version: 1.0
In-Reply-To: <1552472109.10179.103.camel@mhfsdcap03>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Hi,

On 3/13/19 11:15 AM, Chunfeng Yun wrote:
> Hi,
> On Tue, 2019-03-12 at 13:45 +0100, Hans de Goede wrote:
>> Hi,
>> On 12-03-19 04:18, Chunfeng Yun wrote:
>>> Hi,
>>> On Mon, 2019-03-11 at 09:06 +0100, Hans de Goede wrote:
>>>> Hi,
>>>>
>>>> On 11-03-19 06:33, Chunfeng Yun wrote:
>>>>> Hi,
>>>>>
>>>>> On Fri, 2019-03-08 at 13:07 +0100, Hans de Goede wrote:
>>>>>> Hi,
>>>>>>
>>>>>> On 08-03-19 07:13, Chunfeng Yun wrote:
>>>>>>> Add id-gpios, vbus-gpios, vbus-supply and pinctrl properties for
>>>>>>> usb-b-connector
>>>>>>>
>>>>>>> Signed-off-by: Chunfeng Yun <chunfeng.yun@mediatek.com>
>>>>>>> ---
>>>>>>>      .../devicetree/bindings/connector/usb-connector.txt    | 10 ++++++++++
>>>>>>>      1 file changed, 10 insertions(+)
>>>>>>>
>>>>>>> diff --git a/Documentation/devicetree/bindings/connector/usb-connector.txt b/Documentation/devicetree/bindings/connector/usb-connector.txt
>>>>>>> index a9a2f2fc44f2..7a07b0f4f973 100644
>>>>>>> --- a/Documentation/devicetree/bindings/connector/usb-connector.txt
>>>>>>> +++ b/Documentation/devicetree/bindings/connector/usb-connector.txt
>>>>>>> @@ -17,6 +17,16 @@ Optional properties:
>>>>>>>      - self-powered: Set this property if the usb device that has its own power
>>>>>>>        source.
>>>>>>>      
>>>>>>> +Optional properties for usb-b-connector:
>>>>>>> +- id-gpios: gpio for USB ID pin.
>>>>>>
>>>>>> What about boards where the ID pin is *not* connected to a GPIO,
>>>>>> but e.g. to a special pin on the PMIC which can also detect
>>>>>> an ACA adapter ? Currently this case is handled by extcon
>>>>>> drivers, but we have no way to set e.g. vbus-supply for the
>>>>>> connector. Maybe in this case the usb-connector node should
>>>>>> be a child of the PMIC node ?
>>>>> Yes, it would be, PMIC is in charger of detecting the status of ID pin
>>>>
>>>> Ok, then I think this should be documented too.
>>>>
>>>>>> And in many cases there also is a mux to switch the datalines
>>>>>> between the host and device(gadget) controllers, how should
>>>>>> that be described in this model?  See the new usb-role-switch
>>>>>> code under drivers/usb/roles
>>>>>>
>>>>>> In some cases the mux is controlled through a gpio, so we
>>>>>> may want to add a "mux-gpios" here in which case we also
>>>>>> need to define what 0/1 means.
>>>>> I'm not sure, the mux seems not belong to this connector,
>>>>> and may need another driver to register usb-role-switch,
>>>>> similar to:
>>>>>
>>>>> [v2,2/2] usb: typec: add typec switch via GPIO control
>>>>> https://patchwork.kernel.org/patch/10834327/
>>>>
>>>> Right the mux/role-switch will need a driver, but the "owner"
>>>> of the usb_connector, e.g. the PMIC or the owner of the
>>>> id GPIO pin needs to know which device is the role-switch so
>>>> that it can set the role correctly based on the id-pin.
>>>>
>>>> Your binding already contains Vbus info, allowing the owner
>>>> of the usb_connector to enable/disable Vbus based on the id-pin,
>>>> but the owner will also be responsible for setting the role-switch.
>>> In this patch series, I make usb_connector driver enable/disable Vbus,
>>> but not the parent(USB controller) of usb_connector which registers a
>>> usb-role-switch, which way do you think it is better?
>>
>> IMHO there should not be a driver for the usb_connector child-node,
>> only for the parent-device of that child-node.
> If so, each USB controller driver should process this special case by
> itself when use a gpio to detect the status of ID pin, it's not a
> friendly way. And it's easy by using extcon-usb-gpio driver before, so
> we also want to provide a simple way when support usb_connector, no
> matter what method we choose.
> 
> Ideally, the only one thing that USB controller driver need to do, just
> register a usb-role-switch, and leave decision when switch the role to
> other drivers, such as type-C, PMIC/charger, also include gpio
> 
>>
>> There are going to be too many specific setups surrounding PMICs to
>> be able to do a single generic driver for the connector.
>>
> Fortunately, these patches only focus on the simplest gpio driven role
> switch :)

In that case there should be an extra compatible added to the
usb_b_connector node to which the "simple gpio" driver binds,
so that not all devicetree-s declaring a usb_b_connector child-node
automatically get that driver bound, and the gpio / vbus
properties should be mandatory properties for usb_b_connector
child nodes declaring the extra compatible, rather then being
optional for the generic compatible.

Regards,

Hans