Re: PROBLEM: iio: dummy: Oops after mkdir

[Lars-Peter Clausen <lars@metafoo.de>]

On 03/09/2017 07:08 PM, Miguel Robles wrote:
> On Thu, Mar 09, 2017 at 09:40:49AM -0800, Alison Schofield wrote:
>> On Thu, Mar 09, 2017 at 09:36:54AM -0800, Alison Schofield wrote:
>>> On Thu, Mar 09, 2017 at 04:59:39PM +0100, miguel.robles@farole.net wrote:
>>>> On 2017-03-09 16:41, Daniel Baluta wrote:
>>>>> On Thu, Mar 9, 2017 at 4:34 PM, Miguel Robles <miguel.robles@farole.net>
>>>>> wrote:
>>>>>> Hello all,
[...]
>>>>>> CONFIG_IIO_CONFIGFS=m
>>>>>
>>>>> Do you also have CONFIG_IIO_SW_DEVICE right?
>>>>
>>>> Yes, and the value is:  CONFIG_IIO_SW_DEVICE=m.
>>>>
>>>> Thank you,
>>>> Miguel
>>>>>
>>>>> Will try to have a look at this asap.
>>>>>
>>>>>
>>>>> thanks,
>>>>> Daniel.
>>>
>>> Daniel,
>>> You probably saw, but just in case...
>>> I verified that the process, including the above mkdir, is working on
>>> 4.10-rc3.  (So, I'm relieved the Outreachy applicants won't start
>>> hitting it.)
>>> alisons
>>
>> I see Lars' patch! 
>> It worked for me because I used CONFIG_IIO_CONFIGFS built-in.
>> alisons
>>
> I applied Lar's patch and now the command is working fine!
> Even if I do not understand very well the patch. I mean, I always declared
> CONFIG_IIO_CONFIGFS, so the code should be normally always compiled.

What the kernel build system does is when the symbol is select as built-in
(=y) it does `#define CONFIG_CONFIGFS_FS`. When the symbol is selected as a
module (=m) it does `#define CONFIG_CONFIGFS_FS_MODULE`. The IS_ENABLED()
macro basically checks for both and evaluates to true if either of them is
defined.