Re: [PATCH 3/4] watchdog: mt8192: add wdt support

[Crystal Guo]

On Fri, 2020-07-24 at 06:00 +0800, Matthias Brugger wrote:
> 
> On 23/07/2020 11:07, Seiya Wang wrote:
> > From: Crystal Guo <crystal.guo@xxxxxxxxxxxx>
> > 
> > add driver setting to support mt8192 wdt
> > 
> > Signed-off-by: Crystal Guo <crystal.guo@xxxxxxxxxxxx>
> > ---
> >   drivers/watchdog/mtk_wdt.c | 5 +++++
> >   1 file changed, 5 insertions(+)
> > 
> > diff --git a/drivers/watchdog/mtk_wdt.c b/drivers/watchdog/mtk_wdt.c
> > index d6a6393f609d..ad23596170af 100644
> > --- a/drivers/watchdog/mtk_wdt.c
> > +++ b/drivers/watchdog/mtk_wdt.c
> > @@ -76,6 +76,10 @@ static const struct mtk_wdt_data mt8183_data = {
> >   	.toprgu_sw_rst_num = MT8183_TOPRGU_SW_RST_NUM,
> >   };
> >   
> > +static const struct mtk_wdt_data mt8192_data = {
> > +	.toprgu_sw_rst_num = 23,
> 
> Should be defined in include/dt-bindings/reset-controller/mt8192-resets.h

mt8xxx-resets.h actually just used to define TOPRGU_SW_RST_NUM, may be
it's more easier to maintain by define it at OF data directly, instead
of resubmit the mt8xxx-reset.h and add
"include/dt-bindings/reset-controller/mt8xxx-resets.h" at mtk_wdt.c.

Regards
Crystal Guo

> 
> > +};
> > +
> >   static int toprgu_reset_update(struct reset_controller_dev *rcdev,
> >   			       unsigned long id, bool assert)
> >   {
> > @@ -322,6 +326,7 @@ static const struct of_device_id mtk_wdt_dt_ids[] = {
> >   	{ .compatible = "mediatek,mt2712-wdt", .data = &mt2712_data },
> >   	{ .compatible = "mediatek,mt6589-wdt" },
> >   	{ .compatible = "mediatek,mt8183-wdt", .data = &mt8183_data },
> > +	{ .compatible = "mediatek,mt8192-wdt", .data = &mt8192_data },
> >   	{ /* sentinel */ }
> >   };
> >   MODULE_DEVICE_TABLE(of, mtk_wdt_dt_ids);
> >