Re: [PATCH v2] ARM: multi_v7_defconfig: Enable OMAP watchdog support

[Tony Lindgren]

* Julien Panis <jpanis@xxxxxxxxxxxx> [230719 10:09]:
> On 7/18/23 16:58, Julien Panis wrote:
> > Increase build and test coverage by enabling support for OMAP watchdog,
> > as used on TI OMAP based boards.
> > 
> > The watchdog timer is an upward counter capable of generating a pulse on
> > the reset pin and an interrupt to the device system modules following an
> > overflow condition.
> > 
> > Signed-off-by: Julien Panis <jpanis@xxxxxxxxxxxx>

Looks good to me:

Reviewed-by: Tony Lindgren <tony@xxxxxxxxxxx>

> Maybe this patch should not be applied actually. I have 2 questions:
> 
> [Q1] Using the following cmd sequence leads to a modified 'multi_v7_defconfig' file:
>       'make multi_v7_defconfig'
>       'make savedefconfig'
>       'mv defconfig arch/arm/configs/multi_v7_defconfig'
> ...even without modifying CONFIG_OMAP_WATCHDOG flag.
> I guess it's due to modifications in various Kconfig files (dependencies for instance).
> And perhaps it's also due to previous modifications of 'multi_v7_defconfig' file that
> were not done by using 'make savedefconfig' (?)
> How should I handle that for this patch ? This v2 has been created by modifying
> 'multi_v7_defconfig' file manually. Using 'make savedefconfig' would be cleaner,
> but as a result many flags would be re-organized whereas the commit intends to
> enable 1 flag only.

After make savedefconfig you can take a look where the new option got placed
and then throw away the changes and add the entry manually :)

> [Q2] I would like to add another flag in order to enable CONFIG_RTC_DRV_OMAP.
> Is it better grouping CONFIG_OMAP_WATCHDOG and CONFIG_RTC_DRV_OMAP in
> a single commit ? What's recommended ?

Adding both is fine for the defconfig change, just try to place them where
they would end up after savedefconfig to avoid it getting more out of sync.

Regards,

Tony