Re: [PATCH] i2c: core: helper function to detect slave mode

From: Luis Oliveira <Luis.Oliveira@synopsys.com>
To: Andy Shevchenko <andriy.shevchenko@linux.intel.com>,
	Vladimir Zapolskiy <vz@mleia.com>,
	Andy Shevchenko <andy.shevchenko@gmail.com>
Cc: Luis Oliveira <Luis.Oliveira@synopsys.com>,
	Wolfram Sang <wsa@the-dreams.de>,
	Rob Herring <robh+dt@kernel.org>,
	Mark Rutland <mark.rutland@arm.com>,
	Jarkko Nikula <jarkko.nikula@linux.intel.com>,
	"Mika Westerberg" <mika.westerberg@linux.intel.com>,
	<linux-i2c@vger.kernel.org>,
	devicetree <devicetree@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	<Ramiro.Oliveira@synopsys.com>,
	Joao Pinto <Joao.Pinto@synopsys.com>,
	<CARLOS.PALMINHA@synopsys.com>
Subject: Re: [PATCH] i2c: core: helper function to detect slave mode
Date: Mon, 16 Jan 2017 10:32:36 +0000	[thread overview]
Message-ID: <7fdf5d3c-d0ea-ec45-6b18-4573fff6dd11@synopsys.com> (raw)
In-Reply-To: <1484240482.2133.92.camel@linux.intel.com>

On 12-Jan-17 17:01, Andy Shevchenko wrote:
> On Sat, 2017-01-07 at 03:24 +0200, Vladimir Zapolskiy wrote:
>> On 01/07/2017 02:19 AM, Andy Shevchenko wrote:
>>> On Sat, Jan 7, 2017 at 1:43 AM, Vladimir Zapolskiy <vz@mleia.com>
>>> wrote:
>>>> On 01/07/2017 12:45 AM, Andy Shevchenko wrote:
> 
>>>>> +             }
>>>>>>> +     } else if (IS_BUILTIN(CONFIG_ACPI) &&
>>>>>>> ACPI_HANDLE(dev)) {
>>>>>>> +             dev_dbg(dev, "ACPI slave is not supported
>>>>>>> yet\n");
>>>>>>> +     }
>>>>>>
>>>>>> If so, then it might be better to drop else-if stub for now.
>>>>>
>>>>> Please, don't.
>>>>>
>>>>
>>>> Why do you ask for this stub to be added?
>>>
>>> 1. Exactly the reason you asked above. Here is the code which has
>>> built differently on different platforms. x86 usually is not using
>>> CONFIG_OF, ARM doesn't ACPI (versus ARM64). Check GPIO library for
>>> existing examples.
>>
>> From the context by the stub I mean dev_dbg() in
>> i2c_slave_mode_detect()
>> function, I don't see a connection to GPIO library, please clarify.
> 
> I agree that is not good proof for using IS_ENABLED/IS_BUILTIN macro.

I can prepare a V3 and remove it if that's the decision.

> 
>>> 2. We might add that support later, but here is again, just no-op.
>>>
>>> So, what is your strong argument here against that?
>>
>> When the support is ready for ACPI case, you'll remove the added
>> dev_dbg(), and I don't see a good point by adding it temporarily.
> 
> It would remind me to look at it at some point.
> 
>> What is wrong with the approach of adding the ACPI case handling
>> branch when it is ready and remove any kind of stubs right now?
> 
> I will not object. Here is maintainer, let him speak.
> 
>> On ACPI platforms the function returns 'false' always, will the
>> function work correctly (= corresponding to its description) as is?
> 
> Yes.
> 