linux-kernel - Re: [PATCH v5 4/7] dt-bindings: input: matrix_keypad

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <Z74BOePro-J4hSTD@google.com>
Date: Tue, 25 Feb 2025 09:43:21 -0800
From: Dmitry Torokhov <dmitry.torokhov@...il.com>
To: Markus Burri <markus.burri@...com>
Cc: Manuel Traut <manuel.traut@...com>, linux-kernel@...r.kernel.org,
	Rob Herring <robh@...nel.org>,
	Krzysztof Kozlowski <krzk+dt@...nel.org>,
	Conor Dooley <conor+dt@...nel.org>,
	Marek Vasut <marek.vasut@...il.com>, linux-input@...r.kernel.org,
	devicetree@...r.kernel.org
Subject: Re: [PATCH v5 4/7] dt-bindings: input: matrix_keypad - add missing
 property

On Tue, Feb 25, 2025 at 11:16:07AM +0100, Markus Burri wrote:
> On Mon, Feb 24, 2025 at 10:46:09PM -0800, Dmitry Torokhov wrote:
> > On Wed, Feb 19, 2025 at 05:42:00PM +0100, Manuel Traut wrote:
> > > On Fri, Jan 10, 2025 at 06:49:03AM +0100, Markus Burri wrote:
> > > > The property is implemented in the driver but not described in dt-bindings.
> > > > Add missing property 'gpio-activelow' to DT schema.
> > > > 
> > > > Signed-off-by: Markus Burri <markus.burri@...com>
> > > 
> > > Reviewed-by: Manuel Traut <manuel.traut@...com>
> > > 
> > > > ---
> > > >  .../devicetree/bindings/input/gpio-matrix-keypad.yaml       | 6 ++++++
> > > >  1 file changed, 6 insertions(+)
> > > > 
> > > > diff --git a/Documentation/devicetree/bindings/input/gpio-matrix-keypad.yaml b/Documentation/devicetree/bindings/input/gpio-matrix-keypad.yaml
> > > > index 0f348b9..9c91224 100644
> > > > --- a/Documentation/devicetree/bindings/input/gpio-matrix-keypad.yaml
> > > > +++ b/Documentation/devicetree/bindings/input/gpio-matrix-keypad.yaml
> > > > @@ -40,6 +40,12 @@ properties:
> > > >      type: boolean
> > > >      description: Do not enable autorepeat feature.
> > > >  
> > > > +  gpio-activelow:
> > > > +    type: boolean
> > > > +    deprecated: true
> > > > +    description:
> > > > +      The GPIOs are low active (deprecated).
> > > > +      Use the GPIO-flags in gpio controller instead of this property.
> > 
> > No, we unfortunately can not rely on GPIO-flags. This is not how driver
> > works: current driver behavior is to force GPIOs as active high if the
> > property is missing and ignore polarity specified in GPIO property.
> 
> So you mean this property is still valid?
> Current implementation toggle the GPIO flags to low active if the property and the GPIO flags are different.
> So in case of missing property (false) the settings from GPIO flags will be used.
> In case of gpio-activelow (true) GPIO flags are set to low active.

The driver does:

	active_low = device_property_read_bool(&pdev->dev, "gpio-activelow");

	...

		if (active_low ^ gpiod_is_active_low(keypad->col_gpios[i]))
			gpiod_toggle_active_low(keypad->col_gpios[i]);

f the property is present active_low is true. If it is absent active_low
is false. And the code below, if GPIO polarity setting from DT disagree
with active_low value it will flip GPIO polarity.

You probably think that gpiod_toggle_active_low() sets line as active
low, when in fact it flips to opposite setting:

void gpiod_toggle_active_low(struct gpio_desc *desc)
{
	VALIDATE_DESC_VOID(desc);
	change_bit(FLAG_ACTIVE_LOW, &desc->flags);

	^^^ flip, not set or clear.

	gpiod_line_state_notify(desc, GPIO_V2_LINE_CHANGED_CONFIG);
}

> 
> So I will remove the deprecated flag and change the description to:
> "Force GPIO flags to active low. Use GPIO flags if property is missing."
> OK?

No. It should say:

"Force GPIO polarity to active low. In the absence of this property
GPIOs are treated as active high."

Thanks.

-- 
Dmitry