The max98927 codec on some devices (i.e. Xiaomi Mi A2 Lite phone) require
hardware-resetting the codec by driving a reset-gpio. This series add
support for it through an optional reset-gpios property.
Alejandro Tafalla (2):
ASoC: max98927: Handle reset gpio when probing i2c
dt-bindings: sound: max98927: Add reset-gpios optional property
.../devicetree/bindings/sound/max9892x.txt | 3 +++
sound/soc/codecs/max98927.c | 14 ++++++++++++++
2 files changed, 17 insertions(+)
--
2.33.0
From: Alejandro Tafalla <[email protected]>
Drive the reset gpio if defined in the DTS node.
Signed-off-by: Alejandro Tafalla <[email protected]>
---
sound/soc/codecs/max98927.c | 14 ++++++++++++++
1 file changed, 14 insertions(+)
diff --git a/sound/soc/codecs/max98927.c b/sound/soc/codecs/max98927.c
index 8b206ee77709..daf06b503433 100644
--- a/sound/soc/codecs/max98927.c
+++ b/sound/soc/codecs/max98927.c
@@ -868,6 +868,7 @@ static int max98927_i2c_probe(struct i2c_client *i2c,
int ret = 0, value;
int reg = 0;
struct max98927_priv *max98927 = NULL;
+ struct gpio_desc *reset_gpio;
max98927 = devm_kzalloc(&i2c->dev,
sizeof(*max98927), GFP_KERNEL);
@@ -898,6 +899,19 @@ static int max98927_i2c_probe(struct i2c_client *i2c,
return ret;
}
+ reset_gpio
+ = devm_gpiod_get_optional(&i2c->dev, "reset", GPIOD_OUT_LOW);
+ if (IS_ERR(reset_gpio)) {
+ ret = PTR_ERR(reset_gpio);
+ return dev_err_probe(&i2c->dev, ret, "failed to request GPIO reset pin");
+ }
+
+ if (reset_gpio) {
+ usleep_range(8000, 10000);
+ gpiod_set_value_cansleep(reset_gpio, 1);
+ usleep_range(1000, 5000);
+ }
+
/* Check Revision ID */
ret = regmap_read(max98927->regmap,
MAX98927_R01FF_REV_ID, ®);
--
2.33.0
From: Alejandro Tafalla <[email protected]>
Add the reset-gpios as an optional property because some devices might
not need it to work properly.
Signed-off-by: Alejandro Tafalla <[email protected]>
---
Documentation/devicetree/bindings/sound/max9892x.txt | 3 +++
1 file changed, 3 insertions(+)
diff --git a/Documentation/devicetree/bindings/sound/max9892x.txt b/Documentation/devicetree/bindings/sound/max9892x.txt
index f6171591ddc6..98cb9ba5b328 100644
--- a/Documentation/devicetree/bindings/sound/max9892x.txt
+++ b/Documentation/devicetree/bindings/sound/max9892x.txt
@@ -30,6 +30,9 @@ Required properties:
- reg : the I2C address of the device for I2C
+Optional properties:
+ - reset-gpios : GPIO to reset the device
+
Example:
codec: max98927@3a {
--
2.33.0
On Fri, Sep 3, 2021 at 4:51 AM Alejandro <[email protected]> wrote:
>
> The max98927 codec on some devices (i.e. Xiaomi Mi A2 Lite phone) require
requires
> hardware-resetting the codec by driving a reset-gpio. This series add
adds
> support for it through an optional reset-gpios property.
Where is the changelog?
>
> Alejandro Tafalla (2):
> ASoC: max98927: Handle reset gpio when probing i2c
> dt-bindings: sound: max98927: Add reset-gpios optional property
>
> .../devicetree/bindings/sound/max9892x.txt | 3 +++
> sound/soc/codecs/max98927.c | 14 ++++++++++++++
> 2 files changed, 17 insertions(+)
>
> --
> 2.33.0
>
--
With Best Regards,
Andy Shevchenko
On Fri, Sep 3, 2021 at 4:51 AM Alejandro <[email protected]> wrote:
>
> From: Alejandro Tafalla <[email protected]>
>
> Drive the reset gpio if defined in the DTS node.
...
> + reset_gpio
> + = devm_gpiod_get_optional(&i2c->dev, "reset", GPIOD_OUT_LOW);
> + if (IS_ERR(reset_gpio)) {
> + ret = PTR_ERR(reset_gpio);
> + return dev_err_probe(&i2c->dev, ret, "failed to request GPIO reset pin");
Not sure why my comments have been ignored here.
> + }
--
With Best Regards,
Andy Shevchenko
On 03/09/2021 04:49, Alejandro wrote:
> From: Alejandro Tafalla <[email protected]>
>
> Drive the reset gpio if defined in the DTS node.
>
> Signed-off-by: Alejandro Tafalla <[email protected]>
> ---
> sound/soc/codecs/max98927.c | 14 ++++++++++++++
> 1 file changed, 14 insertions(+)
>
> diff --git a/sound/soc/codecs/max98927.c b/sound/soc/codecs/max98927.c
> index 8b206ee77709..daf06b503433 100644
> --- a/sound/soc/codecs/max98927.c
> +++ b/sound/soc/codecs/max98927.c
> @@ -868,6 +868,7 @@ static int max98927_i2c_probe(struct i2c_client *i2c,
> int ret = 0, value;
> int reg = 0;
> struct max98927_priv *max98927 = NULL;
> + struct gpio_desc *reset_gpio;
>
> max98927 = devm_kzalloc(&i2c->dev,
> sizeof(*max98927), GFP_KERNEL);
> @@ -898,6 +899,19 @@ static int max98927_i2c_probe(struct i2c_client *i2c,
> return ret;
> }
>
> + reset_gpio
> + = devm_gpiod_get_optional(&i2c->dev, "reset", GPIOD_OUT_LOW);
If this is a 'reset' pin then it's ACTIVE state is when it places the
device to _reset_.
GPIOD_OUT_LOW == Deasserted state of the GPIO line.
If the reset pin should be pulled low for reset (GPIO_ACTIVE_LOW) and
you want the device initially in reset then you need GPIOD_OUT_HIGH,
because:
GPIOD_OUT_HIGH == Asserted state of the GPIO line.
Same goes for the gpiod_set_value_cansleep():
0 - deasserted
1 = asserted
and this all depends on how the gpio is defined in DT
(GPIO_ACTIVE_LOW/HIGH), which depends on how the documentation refers to
the pin...
reset pin:
low to keep the device in reset, high to release it from reset:
GPIO_ACTIVE_LOW
gpiod_set_value_cansleep(0) to enable
gpiod_set_value_cansleep(1) to disable
enable pin:
high to enable the part, low to disable
GPIO_ACTIVE_HIGH
gpiod_set_value_cansleep(1) to enable
gpiod_set_value_cansleep(0) to disable
In both cases
electrical 0: reset/disable
electrical 1: enable
> + if (IS_ERR(reset_gpio)) {
> + ret = PTR_ERR(reset_gpio);
> + return dev_err_probe(&i2c->dev, ret, "failed to request GPIO reset pin");
> + }
> +
> + if (reset_gpio) {
> + usleep_range(8000, 10000);
> + gpiod_set_value_cansleep(reset_gpio, 1);
> + usleep_range(1000, 5000);
> + }
> +
You might want to put the device to reset on remove at minimum.
> /* Check Revision ID */
> ret = regmap_read(max98927->regmap,
> MAX98927_R01FF_REV_ID, ®);
>
--
Péter
On 3/9/21 10:16 Andy Shevchenko wrote:
> On Fri, Sep 3, 2021 at 4:51 AM Alejandro <[email protected]> wrote:
> > The max98927 codec on some devices (i.e. Xiaomi Mi A2 Lite phone) require
>
> requires
>
> > hardware-resetting the codec by driving a reset-gpio. This series add
>
> adds
>
> > support for it through an optional reset-gpios property.
>
> Where is the changelog?
I'll fix the typos and add previous changelogs in the next version.
Thank you.
--
Alejandro Tafalla
On viernes, 3 de septiembre de 2021 10:18:14 (CEST) Andy Shevchenko wrote:
> On Fri, Sep 3, 2021 at 4:51 AM Alejandro <[email protected]> wrote:
> > From: Alejandro Tafalla <[email protected]>
> >
> > Drive the reset gpio if defined in the DTS node.
>
> ...
>
> > + reset_gpio
> > + = devm_gpiod_get_optional(&i2c->dev, "reset",
> > GPIOD_OUT_LOW); + if (IS_ERR(reset_gpio)) {
> > + ret = PTR_ERR(reset_gpio);
> > + return dev_err_probe(&i2c->dev, ret, "failed to request
> > GPIO reset pin");
> Not sure why my comments have been ignored here.
>
> > + }
>
> --
> With Best Regards,
> Andy Shevchenko
Sorry, I misread your suggestion and didn't notice PTR_ERR was also in the
same line.
On Fri, 03 Sep 2021 03:49:51 +0200, Alejandro wrote:
> From: Alejandro Tafalla <[email protected]>
>
> Add the reset-gpios as an optional property because some devices might
> not need it to work properly.
>
> Signed-off-by: Alejandro Tafalla <[email protected]>
> ---
> Documentation/devicetree/bindings/sound/max9892x.txt | 3 +++
> 1 file changed, 3 insertions(+)
>
Please add Acked-by/Reviewed-by tags when posting new versions. However,
there's no need to repost patches *only* to add the tags. The upstream
maintainer will do that for acks received on the version they apply.
If a tag was not added on purpose, please state why and what changed.
On 3/9/21 19:41 Rob Herring wrote:
> On Fri, 03 Sep 2021 03:49:51 +0200, Alejandro wrote:
> > From: Alejandro Tafalla <[email protected]>
> >
> > Add the reset-gpios as an optional property because some devices might
> > not need it to work properly.
> >
> > Signed-off-by: Alejandro Tafalla <[email protected]>
> > ---
> >
> > Documentation/devicetree/bindings/sound/max9892x.txt | 3 +++
> > 1 file changed, 3 insertions(+)
>
> Please add Acked-by/Reviewed-by tags when posting new versions. However,
> there's no need to repost patches *only* to add the tags. The upstream
> maintainer will do that for acks received on the version they apply.
>
> If a tag was not added on purpose, please state why and what changed.
Okay. I was gonna send another version so I'll add it there.
--
Alejandro Tafalla
On 3/9/21 11:20 P?ter Ujfalusi wrote:
>
> If this is a 'reset' pin then it's ACTIVE state is when it places the
> device to _reset_.
> GPIOD_OUT_LOW == Deasserted state of the GPIO line.
>
> If the reset pin should be pulled low for reset (GPIO_ACTIVE_LOW) and
> you want the device initially in reset then you need GPIOD_OUT_HIGH,
> because:
> GPIOD_OUT_HIGH == Asserted state of the GPIO line.
>
> Same goes for the gpiod_set_value_cansleep():
> 0 - deasserted
> 1 = asserted
>
> and this all depends on how the gpio is defined in DT
> (GPIO_ACTIVE_LOW/HIGH), which depends on how the documentation refers to
> the pin...
>
> reset pin:
> low to keep the device in reset, high to release it from reset:
> GPIO_ACTIVE_LOW
> gpiod_set_value_cansleep(0) to enable
> gpiod_set_value_cansleep(1) to disable
>
>
> enable pin:
> high to enable the part, low to disable
> GPIO_ACTIVE_HIGH
> gpiod_set_value_cansleep(1) to enable
> gpiod_set_value_cansleep(0) to disable
>
> In both cases
> electrical 0: reset/disable
> electrical 1: enable
I'll change it to be consistent in the next version. Thank you for the
explanation.
> > + if (IS_ERR(reset_gpio)) {
> > + ret = PTR_ERR(reset_gpio);
> > + return dev_err_probe(&i2c->dev, ret, "failed to request
GPIO reset
> > pin"); + }
> > +
> > + if (reset_gpio) {
> > + usleep_range(8000, 10000);
> > + gpiod_set_value_cansleep(reset_gpio, 1);
> > + usleep_range(1000, 5000);
> > + }
> > +
>
> You might want to put the device to reset on remove at minimum.
Okay, thanks.
--
Alejandro Tafalla