LEDs next branch

LEDs next branch

[Dan Murphy]

Jacek

Did you see/fix this on the led-next branch?

I saw this when booting trying to fix the lm3532

[    8.651461] led_class: exports duplicate symbol 
devm_led_classdev_register_ext (owned by kernel)

[    8.730063] led_class: exports duplicate symbol 
devm_led_classdev_register_ext (owned by kernel)


Dan

Re: LEDs next branch

[Jacek Anaszewski]

Hi Dan,

On 7/30/19 8:40 PM, Dan Murphy wrote:
> Jacek
> 
> Did you see/fix this on the led-next branch?
> 
> I saw this when booting trying to fix the lm3532
> 
> [    8.651461] led_class: exports duplicate symbol
> devm_led_classdev_register_ext (owned by kernel)
> 
> [    8.730063] led_class: exports duplicate symbol
> devm_led_classdev_register_ext (owned by kernel)

I've just tried next-20190730 and haven't noticed anything like that.
I've checked both with driver compiled-in and loaded as a module.

Tested on qemu versatile board, with tweaked versatile-pb.dts
to include lm3532 DT node.

There is one issue currently in next but seems to be irrelevant and
can be fixed with [0].

[0] https://lkml.org/lkml/2019/7/22/720

-- 
Best regards,
Jacek Anaszewski

Re: LEDs next branch

[Dan Murphy]

Jacek

On 7/30/19 3:30 PM, Jacek Anaszewski wrote:
> Hi Dan,
>
> On 7/30/19 8:40 PM, Dan Murphy wrote:
>> Jacek
>>
>> Did you see/fix this on the led-next branch?
>>
>> I saw this when booting trying to fix the lm3532
>>
>> [    8.651461] led_class: exports duplicate symbol
>> devm_led_classdev_register_ext (owned by kernel)
>>
>> [    8.730063] led_class: exports duplicate symbol
>> devm_led_classdev_register_ext (owned by kernel)
> I've just tried next-20190730 and haven't noticed anything like that.
> I've checked both with driver compiled-in and loaded as a module.
>
> Tested on qemu versatile board, with tweaked versatile-pb.dts
> to include lm3532 DT node.
>
> There is one issue currently in next but seems to be irrelevant and
> can be fixed with [0].

Thanks for checking I will see if I can track it down.

Dan


> [0] https://lkml.org/lkml/2019/7/22/720
>

Re: LEDs next branch

[Dan Murphy]

Jacek

On 7/30/19 3:47 PM, Dan Murphy wrote:
> Jacek
>
> On 7/30/19 3:30 PM, Jacek Anaszewski wrote:
>> Hi Dan,
>>
>> On 7/30/19 8:40 PM, Dan Murphy wrote:
>>> Jacek
>>>
>>> Did you see/fix this on the led-next branch?
>>>
>>> I saw this when booting trying to fix the lm3532
>>>
>>> [    8.651461] led_class: exports duplicate symbol
>>> devm_led_classdev_register_ext (owned by kernel)
>>>
>>> [    8.730063] led_class: exports duplicate symbol
>>> devm_led_classdev_register_ext (owned by kernel)
>> I've just tried next-20190730 and haven't noticed anything like that.
>> I've checked both with driver compiled-in and loaded as a module. 
>
[...]

Try configuring the LEDS_CLASS as a built-in module.

-CONFIG_LEDS_CLASS=m
+CONFIG_LEDS_CLASS=y

I am also on for-next

Dan

Re: LEDs next branch

[Dan Murphy]

Jacek

On 7/30/19 4:11 PM, Dan Murphy wrote:
> Jacek
>
> On 7/30/19 3:47 PM, Dan Murphy wrote:
>> Jacek
>>
>> On 7/30/19 3:30 PM, Jacek Anaszewski wrote:
>>> Hi Dan,
>>>
>>> On 7/30/19 8:40 PM, Dan Murphy wrote:
>>>> Jacek
>>>>
>>>> Did you see/fix this on the led-next branch?
>>>>
>>>> I saw this when booting trying to fix the lm3532
>>>>
>>>> [    8.651461] led_class: exports duplicate symbol
>>>> devm_led_classdev_register_ext (owned by kernel)
>>>>
>>>> [    8.730063] led_class: exports duplicate symbol
>>>> devm_led_classdev_register_ext (owned by kernel)
>>> I've just tried next-20190730 and haven't noticed anything like that.
>>> I've checked both with driver compiled-in and loaded as a module. 
>>
> [...]
>
> Try configuring the LEDS_CLASS as a built-in module.
>
> -CONFIG_LEDS_CLASS=m
> +CONFIG_LEDS_CLASS=y
>
> I am also on for-next
>
I figured it out.  I had old modules which had the led_class as a module 
as opposed to a

built in.

Sorry for the noise

Dan


> Dan
>