Re: [PATCH v9 0/8] Generic PHY Framework

[Kishon Vijay Abraham I <kishon-l0cyMroinI0@public.gmane.org>]

Hi,

On Wednesday 03 July 2013 06:50 PM, Felipe Balbi wrote:
> Hi,
>
> On Wed, Jul 03, 2013 at 03:35:39PM +0530, Kishon Vijay Abraham I wrote:
>> On Wednesday 03 July 2013 03:02 PM, Patel, Satish wrote:
>>> Hi Kishon,
>>>
>>>> -----Original Message-----
>>>> From: ABRAHAM, KISHON VIJAY
>>>> Sent: Wednesday, June 26, 2013 5:17 PM
>>>> To: grant.likely@linaro.org; tony@atomide.com; Balbi, Felipe; ABRAHAM,
>>>> KISHON VIJAY; arnd@arndb.de; swarren@nvidia.com;
>>>> sylvester.nawrocki@gmail.com; linux-kernel@vger.kernel.org; linux-
>>>> omap@vger.kernel.org; linux-arm-kernel@lists.infradead.org; linux-
>>>> usb@vger.kernel.org; gregkh@linuxfoundation.org; akpm@linux-
>>>> foundation.org
>>>> Cc: rob.herring@calxeda.com; rob@landley.net; linux@arm.linux.org.uk;
>>>> benoit.cousson@linaro.org; mchehab@redhat.com; cesarb@cesarb.net;
>>>> davem@davemloft.net; Nayak, Rajendra; shawn.guo@linaro.org; Shilimkar,
>>>> Santosh; devicetree-discuss@lists.ozlabs.org; linux-
>>>> doc@vger.kernel.org; Nori, Sekhar; Krishnamoorthy, Balaji T; Cherian,
>>>> George
>>>> Subject: [PATCH v9 0/8] Generic PHY Framework
>>>>
>>>> Added a generic PHY framework that provides a set of APIs for the PHY
>>>> drivers
>>>> to create/destroy a PHY and APIs for the PHY users to obtain a
>>>> reference to
>>>> the PHY with or without using phandle.
>>>>
>>>> This framework will be of use only to devices that uses external PHY
>>>> (PHY
>>>> functionality is not embedded within the controller).
>>>>
>>>> The intention of creating this framework is to bring the phy drivers
>>>> spread
>>>> all over the Linux kernel to drivers/phy to increase code re-use and
>>>> to
>>>> increase code maintainability.
>>>
>>> I would like to use this framework for a smart-card controller connected to a
>>> smart-card phy. I have some questions and would like to get feedback on the same.
>>
>> glad to know that :-)
>>>
>>> I am using “TDA8026" Smartcard PHY from NXP. Here is the link for datasheet
>>> and app note for the same. The smart card controller is inside the TI SoC
>>> I am working with.
>>>
>>> Datasheet :
>>> www.nxp.com/documents/data_sheet/TDA8026.pdf?
>>>
>>> Appnote :
>>> http://www.nxp.com/documents/application_note/AN10724.pdf
>>>
>>> The TI SoC details are not public (yet). I can provide details to you offline.
>>>
>>> Brief about operation:
>>> -	The controller can work with and without a PHY
>>> -	When not using PHY, it is limited to talking to a single
>>> 	smart card. There is also a need to put external de-activation logic
>>> 	on card removal for this case.
>>> -	With a PHY you can use more than one smart card.
>>> -	Phy has 5 slots :  1 for smart card (credit/debit/other card with chip)
>>>        and others for SAM – SIM like modules
>>> - 	Once the PHY is initialized, there are some operations that the controller
>>> 	can request of the PHY like:
>>> 	- Card configurations  - set voltage
>>> 	- Activation of card
>>> 	- ATR – Answer to reset
>>> 	- Warm reset
>>> 	- ADPU exchange
>>> 	- Deactivation ( Normal/Emergency)
>>
>> hmm.. We should think about extending the phy_ops to include these
>> operations (something like phy_smart_card_ops so that other
>> smart_card PHYs will also be able to use it).
>
> let's try to avoid use-case specific additions. set_voltage sounds like
> a regulator thing, but the regulator is controlled through the PHY. I
> guess it makes sense to have a generic phy_set_voltage() call since even
> USB can make use of that.
>
> For card activation, it sounds like phy_init()/phy_shutdown() would
> cover it.
>
> For warm reset perhaps a phy_reset() callback ? Although that could,
> easily, get abused.
>
> For deactivation, that's phy_shutdown().
>
> ATR and ADPU needs more thought, I guess.
>
>>> - 	In the mode when smartcard controller talks directly to the card without the need
>>> 	for a PHY, all the above operations will be carried out by the controller itself
>>>
>>> My current thought process is to make the controller driver provide the user interface
>>> and talk to the PHY using the generic PHY framework you proposed. In the case where there
>>> is no PHY, my idea is to create a "dummy" PHY which uses the controller functionality itself.
>>
>> right. And in the case where you actually have a PHY, create a PHY
>> driver and implement the phy_smart_card_ops and register with the PHY
>> framework.
>
> I would try to avoid that. Otherwise we will have phy_usb_ops,
> phy_sata_ops, phy_network_ops, phy_pci_ops, etc etc etc. It would easily
> blow up.

true. But it's certainly going to be difficult to map certain function
specific ops to the generic ops (just like ATR and ADPU in smart card
case or set pixel for hdmi).

Thanks
Kishon
_______________________________________________
devicetree-discuss mailing list
devicetree-discuss@lists.ozlabs.org
https://lists.ozlabs.org/listinfo/devicetree-discuss