omap watchdog boot status

omap watchdog boot status

[Cor Peters]

Hello everybody

I was looking into an issue with the omap-wdt.c. The  watchdog driver not is 
reporting a different boot status when a reset is being triggered by the 
watchdog.

>From what I gathered, the issue is that in the omap_wdt_probe function,
pdev->dev->platform_data requires to be a reference to the PRM module,
however it has not been set, and I was wondering how this should work in 
an environment that uses the device tree method. (Link to usage:
http://lxr.free-electrons.com/source/drivers/watchdog/omap_wdt.c#L268[1] ).

My questions are as follows:
1) Is my assertion correct that the current method does not work when 
  the driver is being initialised from an device tree instead of a old style 
  board file.
2) If that is the case, what would be the best method of fixing this situation.

Thank you very much in advance.
-- 

Met vriendelijke groet,
With kind regards,
   Cor Peters

Victron Energy B.V.
Koldingweg 9a, 9723HL Groningen
Tel. +31 36 535 9774   Fax. +31 36 531 1666
www.victronenergy.com

--------
[1] http://lxr.free-electrons.com/source/drivers/watchdog/omap_wdt.c#L268

Re: omap watchdog boot status

[Guenter Roeck]

On 11/07/2016 06:26 AM, Cor Peters wrote:
> Hello everybody
>
> I was looking into an issue with the omap-wdt.c. The  watchdog driver not is
> reporting a different boot status when a reset is being triggered by the
> watchdog.
>
>>From what I gathered, the issue is that in the omap_wdt_probe function,
> pdev->dev->platform_data requires to be a reference to the PRM module,
> however it has not been set, and I was wondering how this should work in
> an environment that uses the device tree method. (Link to usage:
> http://lxr.free-electrons.com/source/drivers/watchdog/omap_wdt.c#L268[1] ).
>
> My questions are as follows:
> 1) Is my assertion correct that the current method does not work when
>   the driver is being initialised from an device tree instead of a old style
>   board file.

I think you are correct.

> 2) If that is the case, what would be the best method of fixing this situation.
>
I am not a devicetree expert, but you may have to define a property to point to
the register and bit providing the necessary information. This could possibly
be done with syscon, but I really don't know enough about it to really know for
sure what the best approach would be.

Guenter

Re: omap watchdog boot status

[Cor Peters]

On Monday 07 November 2016 17:12:56 Guenter Roeck wrote:
> On 11/07/2016 06:26 AM, Cor Peters wrote:
> > Hello everybody
> >
> > I was looking into an issue with the omap-wdt.c. The  watchdog driver not is
> > reporting a different boot status when a reset is being triggered by the
> > watchdog.
> >
> >>From what I gathered, the issue is that in the omap_wdt_probe function,
> > pdev->dev->platform_data requires to be a reference to the PRM module,
> > however it has not been set, and I was wondering how this should work in
> > an environment that uses the device tree method. (Link to usage:
> > http://lxr.free-electrons.com/source/drivers/watchdog/omap_wdt.c#L268[1] ).
> >
> > My questions are as follows:
> > 1) Is my assertion correct that the current method does not work when
> >   the driver is being initialised from an device tree instead of a old style
> >   board file.
> 
> I think you are correct.
> 
> > 2) If that is the case, what would be the best method of fixing this situation.
> >
> I am not a devicetree expert, but you may have to define a property to point to
> the register and bit providing the necessary information. This could possibly
> be done with syscon, but I really don't know enough about it to really know for
> sure what the best approach would be.

Thanks, syscon seems like a good solution.
 
> Guenter
> 

Cor