[v1] drm/sitronix/st7571-i2c: Add support for the ST7567 Controller

[PATCH 2/3] drm/sitronix/st7571-i2c: Make the reset GPIO to be optional

Posted by Javier Martinez Canillas 7 months ago

Some Sitronix LCD controllers (such as the ST7567) don't have a reset pin,
so lets relax this in the driver and make the reset GPIO to be optional.

The devm_gpiod_get_optional() helper is similar to devm_gpiod_get(), but
returns NULL when there isn't a reset-gpios property defined in a DT node.

The DT binding schema for "sitronix,st7571" that require a reset GPIO will
enforce the "reset-gpios" to be present, due being a required DT property.
But in the driver itself the property can be made optional if not defined.

Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
---

 drivers/gpu/drm/sitronix/st7571-i2c.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/sitronix/st7571-i2c.c b/drivers/gpu/drm/sitronix/st7571-i2c.c
index eec846892962..73e8db25f895 100644
--- a/drivers/gpu/drm/sitronix/st7571-i2c.c
+++ b/drivers/gpu/drm/sitronix/st7571-i2c.c
@@ -802,15 +802,19 @@ static int st7571_parse_dt(struct st7571_device *st7571)
 	st7571->nlines = dt.vactive.typ;
 	st7571->ncols = dt.hactive.typ;
 
-	st7571->reset = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
+	st7571->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH);
 	if (IS_ERR(st7571->reset))
-		return PTR_ERR(st7571->reset);
+		return dev_err_probe(dev, PTR_ERR(st7571->reset),
+				     "Failed to get reset gpio\n");
 
 	return 0;
 }
 
 static void st7571_reset(struct st7571_device *st7571)
 {
+	if (!st7571->reset)
+		return;
+
 	gpiod_set_value_cansleep(st7571->reset, 1);
 	fsleep(20);
 	gpiod_set_value_cansleep(st7571->reset, 0);
-- 
2.49.0

Re: [PATCH 2/3] drm/sitronix/st7571-i2c: Make the reset GPIO to be optional

Posted by Thomas Zimmermann 6 months, 4 weeks ago

Hi

Am 10.07.25 um 12:24 schrieb Javier Martinez Canillas:
> Some Sitronix LCD controllers (such as the ST7567) don't have a reset pin,
> so lets relax this in the driver and make the reset GPIO to be optional.
>
> The devm_gpiod_get_optional() helper is similar to devm_gpiod_get(), but
> returns NULL when there isn't a reset-gpios property defined in a DT node.
>
> The DT binding schema for "sitronix,st7571" that require a reset GPIO will
> enforce the "reset-gpios" to be present, due being a required DT property.
> But in the driver itself the property can be made optional if not defined.
>
> Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
> ---
>
>   drivers/gpu/drm/sitronix/st7571-i2c.c | 8 ++++++--
>   1 file changed, 6 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/sitronix/st7571-i2c.c b/drivers/gpu/drm/sitronix/st7571-i2c.c
> index eec846892962..73e8db25f895 100644
> --- a/drivers/gpu/drm/sitronix/st7571-i2c.c
> +++ b/drivers/gpu/drm/sitronix/st7571-i2c.c
> @@ -802,15 +802,19 @@ static int st7571_parse_dt(struct st7571_device *st7571)
>   	st7571->nlines = dt.vactive.typ;
>   	st7571->ncols = dt.hactive.typ;
>   
> -	st7571->reset = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
> +	st7571->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH);
>   	if (IS_ERR(st7571->reset))
> -		return PTR_ERR(st7571->reset);
> +		return dev_err_probe(dev, PTR_ERR(st7571->reset),
> +				     "Failed to get reset gpio\n");

There's struct st7571_panel_data. It could store a flag signalling the 
expected behavior.

With more effort the panel_data could store a dedicated parse_dt pointer 
for each panel type. ASAICT the st7567 features a subset of the other 
type. So there might not be much code duplication.

Best regards
Thomas

>   
>   	return 0;
>   }
>   
>   static void st7571_reset(struct st7571_device *st7571)
>   {
> +	if (!st7571->reset)
> +		return;
> +
>   	gpiod_set_value_cansleep(st7571->reset, 1);
>   	fsleep(20);
>   	gpiod_set_value_cansleep(st7571->reset, 0);

-- 
--
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Frankenstrasse 146, 90461 Nuernberg, Germany
GF: Ivo Totev, Andrew Myers, Andrew McDonald, Boudien Moerman
HRB 36809 (AG Nuernberg)

Re: [PATCH 2/3] drm/sitronix/st7571-i2c: Make the reset GPIO to be optional

Posted by Javier Martinez Canillas 6 months, 4 weeks ago

Thomas Zimmermann <tzimmermann@suse.de> writes:

> Hi
>
> Am 10.07.25 um 12:24 schrieb Javier Martinez Canillas:
>> Some Sitronix LCD controllers (such as the ST7567) don't have a reset pin,
>> so lets relax this in the driver and make the reset GPIO to be optional.
>>
>> The devm_gpiod_get_optional() helper is similar to devm_gpiod_get(), but
>> returns NULL when there isn't a reset-gpios property defined in a DT node.
>>
>> The DT binding schema for "sitronix,st7571" that require a reset GPIO will
>> enforce the "reset-gpios" to be present, due being a required DT property.
>> But in the driver itself the property can be made optional if not defined.
>>
>> Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
>> ---
>>
>>   drivers/gpu/drm/sitronix/st7571-i2c.c | 8 ++++++--
>>   1 file changed, 6 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/sitronix/st7571-i2c.c b/drivers/gpu/drm/sitronix/st7571-i2c.c
>> index eec846892962..73e8db25f895 100644
>> --- a/drivers/gpu/drm/sitronix/st7571-i2c.c
>> +++ b/drivers/gpu/drm/sitronix/st7571-i2c.c
>> @@ -802,15 +802,19 @@ static int st7571_parse_dt(struct st7571_device *st7571)
>>   	st7571->nlines = dt.vactive.typ;
>>   	st7571->ncols = dt.hactive.typ;
>>   
>> -	st7571->reset = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
>> +	st7571->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH);
>>   	if (IS_ERR(st7571->reset))
>> -		return PTR_ERR(st7571->reset);
>> +		return dev_err_probe(dev, PTR_ERR(st7571->reset),
>> +				     "Failed to get reset gpio\n");
>
> There's struct st7571_panel_data. It could store a flag signalling the 
> expected behavior.
>

Indeed.

> With more effort the panel_data could store a dedicated parse_dt pointer 
> for each panel type. ASAICT the st7567 features a subset of the other 
> type. So there might not be much code duplication.
>

This is a good idea too. I think that will do that for v2.

Thanks a lot for your feedback!

-- 
Best regards,

Javier Martinez Canillas
Core Platforms
Red Hat

Re: [PATCH 2/3] drm/sitronix/st7571-i2c: Make the reset GPIO to be optional

Posted by Thomas Zimmermann 6 months, 4 weeks ago

Hi

Am 10.07.25 um 12:24 schrieb Javier Martinez Canillas:
> Some Sitronix LCD controllers (such as the ST7567) don't have a reset pin,
> so lets relax this in the driver and make the reset GPIO to be optional.
>
> The devm_gpiod_get_optional() helper is similar to devm_gpiod_get(), but
> returns NULL when there isn't a reset-gpios property defined in a DT node.
>
> The DT binding schema for "sitronix,st7571" that require a reset GPIO will
> enforce the "reset-gpios" to be present, due being a required DT property.
> But in the driver itself the property can be made optional if not defined.
>
> Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
> ---
>
>   drivers/gpu/drm/sitronix/st7571-i2c.c | 8 ++++++--
>   1 file changed, 6 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/gpu/drm/sitronix/st7571-i2c.c b/drivers/gpu/drm/sitronix/st7571-i2c.c
> index eec846892962..73e8db25f895 100644
> --- a/drivers/gpu/drm/sitronix/st7571-i2c.c
> +++ b/drivers/gpu/drm/sitronix/st7571-i2c.c
> @@ -802,15 +802,19 @@ static int st7571_parse_dt(struct st7571_device *st7571)
>   	st7571->nlines = dt.vactive.typ;
>   	st7571->ncols = dt.hactive.typ;
>   
> -	st7571->reset = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
> +	st7571->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH);
>   	if (IS_ERR(st7571->reset))
> -		return PTR_ERR(st7571->reset);
> +		return dev_err_probe(dev, PTR_ERR(st7571->reset),
> +				     "Failed to get reset gpio\n");
>   
>   	return 0;
>   }
>   
>   static void st7571_reset(struct st7571_device *st7571)
>   {
> +	if (!st7571->reset)
> +		return;
> +

My interpretation of this function is that calling it guarantees a 
device reset (or an error). So I'd push this test into the caller.

Best regards
Thomas

>   	gpiod_set_value_cansleep(st7571->reset, 1);
>   	fsleep(20);
>   	gpiod_set_value_cansleep(st7571->reset, 0);

-- 
--
Thomas Zimmermann
Graphics Driver Developer
SUSE Software Solutions Germany GmbH
Frankenstrasse 146, 90461 Nuernberg, Germany
GF: Ivo Totev, Andrew Myers, Andrew McDonald, Boudien Moerman
HRB 36809 (AG Nuernberg)

Re: [PATCH 2/3] drm/sitronix/st7571-i2c: Make the reset GPIO to be optional

Posted by Javier Martinez Canillas 6 months, 4 weeks ago

Thomas Zimmermann <tzimmermann@suse.de> writes:

Hello Thomas,

> Hi
>
> Am 10.07.25 um 12:24 schrieb Javier Martinez Canillas:
>> Some Sitronix LCD controllers (such as the ST7567) don't have a reset pin,
>> so lets relax this in the driver and make the reset GPIO to be optional.
>>
>> The devm_gpiod_get_optional() helper is similar to devm_gpiod_get(), but
>> returns NULL when there isn't a reset-gpios property defined in a DT node.
>>
>> The DT binding schema for "sitronix,st7571" that require a reset GPIO will
>> enforce the "reset-gpios" to be present, due being a required DT property.
>> But in the driver itself the property can be made optional if not defined.
>>
>> Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
>> ---
>>
>>   drivers/gpu/drm/sitronix/st7571-i2c.c | 8 ++++++--
>>   1 file changed, 6 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/sitronix/st7571-i2c.c b/drivers/gpu/drm/sitronix/st7571-i2c.c
>> index eec846892962..73e8db25f895 100644
>> --- a/drivers/gpu/drm/sitronix/st7571-i2c.c
>> +++ b/drivers/gpu/drm/sitronix/st7571-i2c.c
>> @@ -802,15 +802,19 @@ static int st7571_parse_dt(struct st7571_device *st7571)
>>   	st7571->nlines = dt.vactive.typ;
>>   	st7571->ncols = dt.hactive.typ;
>>   
>> -	st7571->reset = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
>> +	st7571->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH);
>>   	if (IS_ERR(st7571->reset))
>> -		return PTR_ERR(st7571->reset);
>> +		return dev_err_probe(dev, PTR_ERR(st7571->reset),
>> +				     "Failed to get reset gpio\n");
>>   
>>   	return 0;
>>   }
>>   
>>   static void st7571_reset(struct st7571_device *st7571)
>>   {
>> +	if (!st7571->reset)
>> +		return;
>> +
>
> My interpretation of this function is that calling it guarantees a 
> device reset (or an error). So I'd push this test into the caller.
>

That's a good point. I'll then do the check in the caller.

Actually... at the end I didn't need a st7571_reset() call for ST7567
since it has its own struct st7571_panel_data .init callback function.

So I can just drop the test for st7571->reset being NULL.

-- 
Best regards,

Javier Martinez Canillas
Core Platforms
Red Hat

Re: [PATCH 2/3] drm/sitronix/st7571-i2c: Make the reset GPIO to be optional

Posted by Marcus Folkesson 7 months ago

On Thu, Jul 10, 2025 at 12:24:34PM +0200, Javier Martinez Canillas wrote:
> Some Sitronix LCD controllers (such as the ST7567) don't have a reset pin,
> so lets relax this in the driver and make the reset GPIO to be optional.
> 
> The devm_gpiod_get_optional() helper is similar to devm_gpiod_get(), but
> returns NULL when there isn't a reset-gpios property defined in a DT node.
> 
> The DT binding schema for "sitronix,st7571" that require a reset GPIO will
> enforce the "reset-gpios" to be present, due being a required DT property.
> But in the driver itself the property can be made optional if not defined.
> 
> Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
> ---
> 
>  drivers/gpu/drm/sitronix/st7571-i2c.c | 8 ++++++--
>  1 file changed, 6 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/gpu/drm/sitronix/st7571-i2c.c b/drivers/gpu/drm/sitronix/st7571-i2c.c
> index eec846892962..73e8db25f895 100644
> --- a/drivers/gpu/drm/sitronix/st7571-i2c.c
> +++ b/drivers/gpu/drm/sitronix/st7571-i2c.c
> @@ -802,15 +802,19 @@ static int st7571_parse_dt(struct st7571_device *st7571)
>  	st7571->nlines = dt.vactive.typ;
>  	st7571->ncols = dt.hactive.typ;
>  
> -	st7571->reset = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
> +	st7571->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH);
>  	if (IS_ERR(st7571->reset))
> -		return PTR_ERR(st7571->reset);
> +		return dev_err_probe(dev, PTR_ERR(st7571->reset),
> +				     "Failed to get reset gpio\n");

devm_gpiod_get_optional() returns -ENOENT when the GPIO is not found,
and that is no error we want to propagage upwards.

Maybe something like this instead:
if (IS_ERR(st7571->reset) && IS_ERR(st7571->reset) != -ENOENT)


Best regards,
Marcus Folkesson

Re: [PATCH 2/3] drm/sitronix/st7571-i2c: Make the reset GPIO to be optional

Posted by Javier Martinez Canillas 7 months ago

Marcus Folkesson <marcus.folkesson@gmail.com> writes:

Hello Marcus,

Thanks for your feedback.

> On Thu, Jul 10, 2025 at 12:24:34PM +0200, Javier Martinez Canillas wrote:
>> Some Sitronix LCD controllers (such as the ST7567) don't have a reset pin,
>> so lets relax this in the driver and make the reset GPIO to be optional.
>> 
>> The devm_gpiod_get_optional() helper is similar to devm_gpiod_get(), but
>> returns NULL when there isn't a reset-gpios property defined in a DT node.
>> 
>> The DT binding schema for "sitronix,st7571" that require a reset GPIO will
>> enforce the "reset-gpios" to be present, due being a required DT property.
>> But in the driver itself the property can be made optional if not defined.
>> 
>> Signed-off-by: Javier Martinez Canillas <javierm@redhat.com>
>> ---
>> 
>>  drivers/gpu/drm/sitronix/st7571-i2c.c | 8 ++++++--
>>  1 file changed, 6 insertions(+), 2 deletions(-)
>> 
>> diff --git a/drivers/gpu/drm/sitronix/st7571-i2c.c b/drivers/gpu/drm/sitronix/st7571-i2c.c
>> index eec846892962..73e8db25f895 100644
>> --- a/drivers/gpu/drm/sitronix/st7571-i2c.c
>> +++ b/drivers/gpu/drm/sitronix/st7571-i2c.c
>> @@ -802,15 +802,19 @@ static int st7571_parse_dt(struct st7571_device *st7571)
>>  	st7571->nlines = dt.vactive.typ;
>>  	st7571->ncols = dt.hactive.typ;
>>  
>> -	st7571->reset = devm_gpiod_get(dev, "reset", GPIOD_OUT_HIGH);
>> +	st7571->reset = devm_gpiod_get_optional(dev, "reset", GPIOD_OUT_HIGH);
>>  	if (IS_ERR(st7571->reset))
>> -		return PTR_ERR(st7571->reset);
>> +		return dev_err_probe(dev, PTR_ERR(st7571->reset),
>> +				     "Failed to get reset gpio\n");
>
> devm_gpiod_get_optional() returns -ENOENT when the GPIO is not found,
> and that is no error we want to propagage upwards.
>
> Maybe something like this instead:
> if (IS_ERR(st7571->reset) && IS_ERR(st7571->reset) != -ENOENT)
>

Are you sure about that? As far as I know, that is exactly the
difference between gpiod_get() and gpiod_get_optional() variants.

From the gpiod_get_optional() function helper kernel-doc [0]:

/**
 * gpiod_get_optional - obtain an optional GPIO for a given GPIO function
 * @dev: GPIO consumer, can be NULL for system-global GPIOs
 * @con_id: function within the GPIO consumer
 * @flags: optional GPIO initialization flags
 *
 * This is equivalent to gpiod_get(), except that when no GPIO was assigned to
 * the requested function it will return NULL. This is convenient for drivers
 * that need to handle optional GPIOs.
 *
 * Returns:
 * The GPIO descriptor corresponding to the function @con_id of device
 * dev, NULL if no GPIO has been assigned to the requested function, or
 * another IS_ERR() code if an error occurred while trying to acquire the GPIO.
 */

while the gpiod_get() kernel-doc says the following:

/**
 * gpiod_get - obtain a GPIO for a given GPIO function
 * @dev:	GPIO consumer, can be NULL for system-global GPIOs
 * @con_id:	function within the GPIO consumer
 * @flags:	optional GPIO initialization flags
 *
 * Returns:
 * The GPIO descriptor corresponding to the function @con_id of device
 * dev, -ENOENT if no GPIO has been assigned to the requested function, or
 * another IS_ERR() code if an error occurred while trying to acquire the GPIO.
 */

[0]: https://elixir.bootlin.com/linux/v6.16-rc5/source/drivers/gpio/gpiolib.c#L4755

>
> Best regards,
> Marcus Folkesson

-- 
Best regards,

Javier Martinez Canillas
Core Platforms
Red Hat

Re: [PATCH 2/3] drm/sitronix/st7571-i2c: Make the reset GPIO to be optional

Posted by Marcus Folkesson 7 months ago

Hello Javier,

On Thu, Jul 10, 2025 at 01:00:41PM +0200, Javier Martinez Canillas wrote:
> >
> > devm_gpiod_get_optional() returns -ENOENT when the GPIO is not found,
> > and that is no error we want to propagage upwards.
> >
> > Maybe something like this instead:
> > if (IS_ERR(st7571->reset) && IS_ERR(st7571->reset) != -ENOENT)
> >
> 
> Are you sure about that? As far as I know, that is exactly the
> difference between gpiod_get() and gpiod_get_optional() variants.
> 
> From the gpiod_get_optional() function helper kernel-doc [0]:
> 
> /**
>  * gpiod_get_optional - obtain an optional GPIO for a given GPIO function
>  * @dev: GPIO consumer, can be NULL for system-global GPIOs
>  * @con_id: function within the GPIO consumer
>  * @flags: optional GPIO initialization flags
>  *
>  * This is equivalent to gpiod_get(), except that when no GPIO was assigned to
>  * the requested function it will return NULL. This is convenient for drivers
>  * that need to handle optional GPIOs.
>  *
>  * Returns:
>  * The GPIO descriptor corresponding to the function @con_id of device
>  * dev, NULL if no GPIO has been assigned to the requested function, or
>  * another IS_ERR() code if an error occurred while trying to acquire the GPIO.
>  */
> 
> while the gpiod_get() kernel-doc says the following:
> 
> /**
>  * gpiod_get - obtain a GPIO for a given GPIO function
>  * @dev:	GPIO consumer, can be NULL for system-global GPIOs
>  * @con_id:	function within the GPIO consumer
>  * @flags:	optional GPIO initialization flags
>  *
>  * Returns:
>  * The GPIO descriptor corresponding to the function @con_id of device
>  * dev, -ENOENT if no GPIO has been assigned to the requested function, or
>  * another IS_ERR() code if an error occurred while trying to acquire the GPIO.
>  */
> 

You are completely righ.

Reviewed-by: Marcus Folkesson <marcus.folkesson@gmail.com>

Best regards,
Marcus Folkesson

Re: [PATCH 2/3] drm/sitronix/st7571-i2c: Make the reset GPIO to be optional

Posted by Javier Martinez Canillas 7 months ago

Marcus Folkesson <marcus.folkesson@gmail.com> writes:

Hello Marcus,

> Hello Javier,
>
> On Thu, Jul 10, 2025 at 01:00:41PM +0200, Javier Martinez Canillas wrote:
>> >
>> > devm_gpiod_get_optional() returns -ENOENT when the GPIO is not found,
>> > and that is no error we want to propagage upwards.
>> >
>> > Maybe something like this instead:
>> > if (IS_ERR(st7571->reset) && IS_ERR(st7571->reset) != -ENOENT)
>> >
>> 
>> Are you sure about that? As far as I know, that is exactly the
>> difference between gpiod_get() and gpiod_get_optional() variants.
>> 
>> From the gpiod_get_optional() function helper kernel-doc [0]:
>> 
>> /**
>>  * gpiod_get_optional - obtain an optional GPIO for a given GPIO function
>>  * @dev: GPIO consumer, can be NULL for system-global GPIOs
>>  * @con_id: function within the GPIO consumer
>>  * @flags: optional GPIO initialization flags
>>  *
>>  * This is equivalent to gpiod_get(), except that when no GPIO was assigned to
>>  * the requested function it will return NULL. This is convenient for drivers
>>  * that need to handle optional GPIOs.
>>  *
>>  * Returns:
>>  * The GPIO descriptor corresponding to the function @con_id of device
>>  * dev, NULL if no GPIO has been assigned to the requested function, or
>>  * another IS_ERR() code if an error occurred while trying to acquire the GPIO.
>>  */
>> 
>> while the gpiod_get() kernel-doc says the following:
>> 
>> /**
>>  * gpiod_get - obtain a GPIO for a given GPIO function
>>  * @dev:	GPIO consumer, can be NULL for system-global GPIOs
>>  * @con_id:	function within the GPIO consumer
>>  * @flags:	optional GPIO initialization flags
>>  *
>>  * Returns:
>>  * The GPIO descriptor corresponding to the function @con_id of device
>>  * dev, -ENOENT if no GPIO has been assigned to the requested function, or
>>  * another IS_ERR() code if an error occurred while trying to acquire the GPIO.
>>  */
>> 
>
> You are completely righ.
>
> Reviewed-by: Marcus Folkesson <marcus.folkesson@gmail.com>
>

Thanks for your review! Do you plan to look at the other patches too ?

-- 
Best regards,

Javier Martinez Canillas
Core Platforms
Red Hat