Re: [Bug]:LAGER: GPIO-KEYS: Warning occurs after unbinding the e6051000.gpio

[Hiep Cao Minh <cm-hiep@jinso.co.jp>]

Hi Linus-san

I am Hiep from Jinzai Solution.
We work for Renesas's drivers development project.

I am sorry to bother you again!
As I have reported you the issue that related to your released patch.
Could you please consider this issue and tell me how to resolve this 
problem?
This issue still happens on Rcar-H2 and Rcar-M2 (Renesas's SOC) at the 
Linux upstream version v4.7-rc2.

Thank you!
Hiep.


On 04/27/2016 02:23 PM, Cao Minh Hiep wrote:
> Hi Linus-san
> Thanks for your reply!
>
> On 04/26/2016 06:20 PM, Linus Walleij wrote:
>> On Mon, Apr 25, 2016 at 11:32 AM, Cao Minh Hiep <cm-hiep@jinso.co.jp> 
>> wrote:
>>
>>> Hello Linus Walleij-san
>>>
>>> We have tested Linux upstream v4.6-rc2 on Renesas's Lager board.
>>> When we tried to unbind the e6051000.gpio, the following warning 
>>> messages
>>> occurs:
>>>
>>> "root@linaro-nano:/sys/bus/platform/drivers/gpio_rcar# echo 
>>> e6051000.gpio >
>>> unbind
>>> [  241.511034] ------------[ cut here ]------------
>>> [  241.525054] WARNING: CPU: 0 PID: 2104 at fs/proc/generic.c:575
>>> remove_proc_entry+0x13c/0x160
>>> [  241.550456] remove_proc_entry: removing non-empty directory 
>>> 'irq/169',
>>> leaking at least '6-0039'
>> Do you mean that you set up a handler in userspace, using the
>> deprecated sysfs ABI and then unbind the module providing the
>> IRQ resource?
>>
> Please see the attach file for procedure test of this case.
> In order to make sure that driver works normally after unbind and 
> rebind the device.
>
>>> And we found a patch between v4.5 and v4.6-rc2 that causing of this 
>>> issue.
>>> The patch is "ff2b1359 gpio: make the gpiochip a real device"
>> It seems the issue is not a bug in the kernel, the issue is that the
>> kernel is warning you about something that was wrong also before
>> but you didn't get a warning for it until now.
>>
>> It is not OK to unbind a driver providing IRQs.
>>
>> It is even unclear if we should even allow modules to provide
>> IRQs because there is no way to handle irqchips going away
>> when it has consumers.
>>
>> Maybe we should just do this?
>>
>> diff --git a/drivers/gpio/gpio-rcar.c b/drivers/gpio/gpio-rcar.c
>> index 3fe8e773d95c..ae5e81358ec5 100644
>> --- a/drivers/gpio/gpio-rcar.c
>> +++ b/drivers/gpio/gpio-rcar.c
>> @@ -562,6 +562,7 @@ static struct platform_driver 
>> gpio_rcar_device_driver = {
>>          .driver         = {
>>                  .name   = "gpio_rcar",
>>                  .of_match_table = of_match_ptr(gpio_rcar_of_table),
>> +               .suppress_bind_attrs = true,
>>          }
>>   };
>>
> I have just tried to patch this patch on v4.6-rc2, then make a test.
> The results is: I could not unbind the device.
>
> "Welcome to Linaro 14.04 (GNU/Linux 4.6.0-rc2-dirty armv7l)
>
>  * Documentation:  https://wiki.linaro.org/
> root@linaro-nano:~# uname -a
> Linux linaro-nano 4.6.0-rc2-dirty #8 SMP Wed Apr 27 13:40:12 JST 2016 
> armv7l armv7l armv7l GNU/Linux
> root@linaro-nano:~# cd /sys/bus/platform/drivers/gpio-keys; ls -d 
> keyboard
> keyboard
> root@linaro-nano:/sys/bus/platform/drivers/gpio-keys# echo keyboard > 
> unbind
> root@linaro-nano:/sys/bus/platform/drivers/gpio-keys# cd
> root@linaro-nano:~#
> root@linaro-nano:~# cd /sys/bus/platform/drivers/gpio_rcar; ls -d 
> e6051000.gpio
> e6051000.gpio
> root@linaro-nano:/sys/bus/platform/drivers/gpio_rcar# echo 
> e6051000.gpio > unbind
> -bash: unbind: Permission denied
>
> root@linaro-nano:/sys/bus/platform/drivers/gpio_rcar# ls
> e6050000.gpio  e6052000.gpio  e6054000.gpio  uevent
> e6051000.gpio  e6053000.gpio  e6055000.gpio
> root@linaro-nano:/sys/bus/platform/drivers/gpio_rcar#
> "
>
> before patching the patch, gpio-rcar devices are shown as below:
>
> "root@linaro-nano:~# cd /sys/bus/platform/drivers/gpio_rcar; ls
> bind           e6051000.gpio  e6053000.gpio  e6055000.gpio  unbind
> e6050000.gpio  e6052000.gpio  e6054000.gpio  uevent
> root@linaro-nano:/sys/bus/platform/drivers/gpio_rcar# "
>
>
> Best Regards,
> Jinso/Cao Minh Hiep.
>