Hi Sudhakar,

On 08.02.2011 10:30, Rajashekhara, Sudhakar wrote:
> Hi,
> 
> On Thu, Feb 03, 2011 at 15:08:39, Christophe Aeschlimann wrote:
>> Hi,
>>
>> I'm working with the arago omapl1 kernel on a custom OMAP-L138 board.
>>
>> http://arago-project.org/git/projects/?p=linux-omapl1.git
>>
>> Recently I made some testing with the PREEMPT_RT patches (that were also
>> applied in a branch in the arago kernel). The Kernel seems to behave
>> correctly but I have issues with the GPIO interface. E.G. I'm using the
>> "gpio-keys" and "matrix-keypad" drivers and they don't behave correctly.
>> The symptoms are :
>>
>> Just after boot, I can press a button attached to a GPIO and get an
>> event through the input subsystem, but after a few presses (few being
>> random) the button doesn't generate an event any-more. (the interrupt is
>> also not reported in /proc/interrupts).
>>
>> By browsing the mailing list archives I saw these old patches for the
>> davinci tree and the omap tree :
>>
>> http://linux.omap.com/pipermail/davinci-linux-open-source/2007-October/004283.html
>>
>> http://marc.info/?l=linux-omap&m=119634477133936&w=2
>>
>> (both are also listed here
>> https://rt.wiki.kernel.org/index.php/CONFIG_PREEMPT_RT_Patch)
>>
>> Of course quite a lot of things changed since 2.6.23 so it's difficult
>> to just adapt the patches (I don't even know if they are still required
>> or if I'm facing another issue).
>>
>> Anyway just wanted to know if someone is having the same issues or if
>> there is a new version of the patches for the latest davinci kernel that
>> I may have missed.
>>
> 
> I downloaded the Linux Kernel from arago which had these patches applied. I
> added support for "gpio-keys-polled" driver and tested it on AM1810 EVM from
> spectrum digital. I did not observe the issue you have reported. I pressed
> the push buttons on the daughter card for about 800 times and I used evtest
> to test.

Thanks for doing some testing on your side.

Any specific reason why you used the gpio-keys-polled driver ? I'm using
the standard gpio-keys driver which is based on a interrupt instead of
polling. As I said I think it's a problem with the interruption mechanism.

I'm defining my buttons with a debounce_interval of 30 in case it
changes anything.

Thanks for your help,

> Regards,
> Sudhakar

Best regards,

Christophe


-- 
Christophe Aeschlimann

Embedded Software Engineer & IT Manager

ACN Advanced Communications Networks S.A.

Rue du Puits-Godet 8a
2000 Neuchâtel, Switzerland

Tél. +41 32 724 74 31

[email protected]
_______________________________________________
Davinci-linux-open-source mailing list
[email protected]
http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source

Reply via email to