[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20250730090605.2824394-1-cy_huang@richtek.com>
Date: Wed, 30 Jul 2025 17:06:05 +0800
From: <cy_huang@...htek.com>
To: <krzk@...nel.org>
CC: <conor+dt@...nel.org>, <cy_huang@...htek.com>,
<devicetree@...r.kernel.org>, <krzk+dt@...nel.org>,
<linux-kernel@...r.kernel.org>, <linux-pm@...r.kernel.org>,
<robh@...nel.org>, <sre@...nel.org>
Subject: Re: [PATCH 1/3] dt-bindings: power: supply: Add Richtek RT9756 smart cap divider charger
> On Wed, Jul 30, 2025 at 09:39:38AM +0200, Krzysztof Kozlowski wrote:
> > On Wed, Jul 30, 2025 at 10:07:16AM +0800, ChiYuan Huang wrote:
> > > On Tue, Jul 29, 2025 at 05:40:32PM +0200, Krzysztof Kozlowski wrote:
> > > > On 29/07/2025 06:21, cy_huang@...htek.com wrote:
> > > > > +
> > > > > + shunt-resistor-micro-ohms:
> > > > > + description: Battery current sense resistor mounted.
> > > > > + default: 2000
> > > > > +
> > > > > +required:
> > > > > + - compatible
> > > > > + - reg
> > > > > + - wakeup-source
> > > >
> > > > Why do you require this? I cannot find any use of it, so maybe I missed
> > > > some change in Linux code (and that's second question like that for
> > > > Richtek, so refer to your other patchsets for contexr).
> > > >
> > >
> > > This will mark the interrupt as wakeup capable.
> > > https://elixir.bootlin.com/linux/v6.16/source/drivers/i2c/i2c-core-of.c#L57
> > > https://elixir.bootlin.com/linux/v6.16/source/drivers/i2c/i2c-core-base.c#L547
> >
> > OK, but this does not explain why this is required. Why it is impossible
> > to make board which uses this PMIC and wires the interrupt in a way it
> > is not waking up the system?
> >
> > To my limited knowledge this should be possible, but what do I know
> > about hardware...
>
> Another question is still valid, although you provided more context - if
> the device is ALWAYS waking up, this is implied by compatible and you do
> not need this property at all.
>
> That would be the first usage of I2C client wakeup flag in drivers, but
> maybe that is how it should be done? You can consult I2C folks on IRC.
Yap, like as your saying. If not declared 'wakeup-source', it only means
if the system supports sleep state, this interrupt won't on-time notify any
event. Actually, not affect its normal functionality.
If your question is should the property 'wakeup-source' be declared as
required, then the answer should be 'No'.
When system supports suspend, to declare this property is just a suggestion.
In next revision, I'll remove 'wakeup-source' from the required property list.
All questions are clarified.
Thx.
Powered by blists - more mailing lists