onewire: w1-gpio: add ext_pullup_enable pin in platform data

PATCH In review Daniel Mack 4 participants 11 messages Jul 25, 2012 – Aug 16, 2012

Subsystems (2)

Device tree bindings W1 dallas's 1-wire bus

lore

Share & tools

lore

Daniel Mack

author

2012-07-25 20:54 UTC #0 lore raw

PATCH 1/2 onewire: w1-gpio: add ext_pullup_enable pin in platform data

In the process of porting boards to devicetree implemenation, we should keep information about external circuitry where they belong - the individual drivers.

This patch adds a way to specify a GPIO to drive the (optional) external pull-up logic, rather than using a function pointer for that.

Signed-off-by Daniel Mack <[email protected]>

Cc Evgeniy Polyakov <[email protected]>

Cc Ville Syrjala <[email protected]>

---
 drivers/w1/masters/w1-gpio.c |   18 +++++++++++++++++-
 include/linux/w1-gpio.h      |    1 +
 2 files changed, 18 insertions(+), 1 deletion(-)

drivers/w1/masters/w1-gpio.c +17 -1

index df600d1..dac88ff 100644
@@ -59,6 +59,13 @@ static int __init w1_gpio_probe(struct platform_device *pdev)
 	if (err)
 		goto free_master;
 
+	if (gpio_is_valid(pdata->ext_pullup_enable_pin)) {
+		err = gpio_request_one(pdata->ext_pullup_enable_pin,
+				       GPIOF_INIT_LOW, "w1 pullup");
+		if (err < 0)
+			goto free_gpio;
+	}
+
 	master->data = pdata;
 	master->read_bit = w1_gpio_read_bit;

@@ -72,15 +79,21 @@ static int __init w1_gpio_probe(struct platform_device *pdev)
 
 	err = w1_add_master_device(master);
 	if (err)
-		goto free_gpio;
+		goto free_gpio_ext_pu;
 
 	if (pdata->enable_external_pullup)
 		pdata->enable_external_pullup(1);
 
+	if (gpio_is_valid(pdata->ext_pullup_enable_pin))
+		gpio_set_value(pdata->ext_pullup_enable_pin, 1);
+
 	platform_set_drvdata(pdev, master);
 
 	return 0;
 
+ free_gpio_ext_pu:
+	if (gpio_is_valid(pdata->ext_pullup_enable_pin))
+		gpio_free(pdata->ext_pullup_enable_pin);
  free_gpio:
 	gpio_free(pdata->pin);
  free_master:

@@ -97,6 +110,9 @@ static int __exit w1_gpio_remove(struct platform_device *pdev)
 	if (pdata->enable_external_pullup)
 		pdata->enable_external_pullup(0);
 
+	if (gpio_is_valid(pdata->ext_pullup_enable_pin))
+		gpio_set_value(pdata->ext_pullup_enable_pin, 0);
+
 	w1_remove_master_device(master);
 	gpio_free(pdata->pin);
 	kfree(master);

include/linux/w1-gpio.h +1 -0

index 3adeff8..065e3ae 100644
@@ -19,6 +19,7 @@ struct w1_gpio_platform_data {
 	unsigned int pin;
 	unsigned int is_open_drain:1;
 	void (*enable_external_pullup)(int enable);
+	unsigned int ext_pullup_enable_pin;
 };
 
 #endif /* _LINUX_W1_GPIO_H */

Signature · 1 line

1.7.10.4

Daniel Mack

author

2012-07-25 20:54 UTC #1 lore raw

PATCH 2/2 onewire: w1-gpio: add DT bindings

This patch add DT bindings to the w1-gpio driver, along with some documentation on how to use them.

Signed-off-by Daniel Mack <[email protected]>

Cc Evgeniy Polyakov <[email protected]>

Cc Ville Syrjala <[email protected]>

---
 Documentation/devicetree/bindings/w1/w1-gpio.txt |   22 +++++++++++
 drivers/w1/masters/w1-gpio.c                     |   46 ++++++++++++++++++++++
 2 files changed, 68 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/w1/w1-gpio.txt

Documentation/devicetree/bindings/w1/w1-gpio.txt new file +22 -0

new file mode 100644
index 0000000..6e09c35
@@ -0,0 +1,22 @@
+w1-gpio devicetree bindings
+
+Required properties:
+
+ - compatible: "w1-gpio"
+ - gpios: one or two GPIO specs:
+		- the first one is used as data I/O pin
+		- the second one is optional. If specified, it is used as
+		  enable pin for an external pin pullup.
+
+Optional properties:
+
+ - linux,open-drain: if specified, the data pin is considered in
+		     open-drain mode.
+
+Examples:
+
+	onewire@0 {
+		compatible = "w1-gpio";
+		gpios = <&gpio 126 0>, <&gpio 105 0>;
+	};
+

drivers/w1/masters/w1-gpio.c +46 -0

70 lines not on this page · raw message

Signature · 1 line

1.7.10.4

Daniel Mack

author

2012-07-30 17:26 UTC #2 lore raw

In reply to Daniel Mack

Ping.

Quoted 79 lines · Daniel Mack

In the process of porting boards to devicetree implemenation, we should keep information about external circuitry where they belong - the individual drivers.

This patch adds a way to specify a GPIO to drive the (optional) external pull-up logic, rather than using a function pointer for that.

Signed-off-by Daniel Mack <[email protected]>

Cc Evgeniy Polyakov <[email protected]>

Cc Ville Syrjala <[email protected]>

---
 drivers/w1/masters/w1-gpio.c |   18 +++++++++++++++++-
 include/linux/w1-gpio.h      |    1 +
 2 files changed, 18 insertions(+), 1 deletion(-)

drivers/w1/masters/w1-gpio.c +17 -1

index df600d1..dac88ff 100644
@@ -59,6 +59,13 @@ static int __init w1_gpio_probe(struct platform_device *pdev)
 	if (err)
 		goto free_master;
 
+	if (gpio_is_valid(pdata->ext_pullup_enable_pin)) {
+		err = gpio_request_one(pdata->ext_pullup_enable_pin,
+				       GPIOF_INIT_LOW, "w1 pullup");
+		if (err < 0)
+			goto free_gpio;
+	}
+
 	master->data = pdata;
 	master->read_bit = w1_gpio_read_bit;

@@ -72,15 +79,21 @@ static int __init w1_gpio_probe(struct platform_device *pdev)
 
 	err = w1_add_master_device(master);
 	if (err)
-		goto free_gpio;
+		goto free_gpio_ext_pu;
 
 	if (pdata->enable_external_pullup)
 		pdata->enable_external_pullup(1);
 
+	if (gpio_is_valid(pdata->ext_pullup_enable_pin))
+		gpio_set_value(pdata->ext_pullup_enable_pin, 1);
+
 	platform_set_drvdata(pdev, master);
 
 	return 0;
 
+ free_gpio_ext_pu:
+	if (gpio_is_valid(pdata->ext_pullup_enable_pin))
+		gpio_free(pdata->ext_pullup_enable_pin);
  free_gpio:
 	gpio_free(pdata->pin);
  free_master:

@@ -97,6 +110,9 @@ static int __exit w1_gpio_remove(struct platform_device *pdev)
 	if (pdata->enable_external_pullup)
 		pdata->enable_external_pullup(0);
 
+	if (gpio_is_valid(pdata->ext_pullup_enable_pin))
+		gpio_set_value(pdata->ext_pullup_enable_pin, 0);
+
 	w1_remove_master_device(master);
 	gpio_free(pdata->pin);
 	kfree(master);

include/linux/w1-gpio.h +1 -0

index 3adeff8..065e3ae 100644
@@ -19,6 +19,7 @@ struct w1_gpio_platform_data {
 	unsigned int pin;
 	unsigned int is_open_drain:1;
 	void (*enable_external_pullup)(int enable);
+	unsigned int ext_pullup_enable_pin;
 };
 
 #endif /* _LINUX_W1_GPIO_H */

Evgeniy Polyakov

Acked-by on v1 1/2

2012-07-30 17:41 UTC #3 lore raw

In reply to Daniel Mack

Sorry for long delay

Quoted 6 lines · Daniel Mack

In the process of porting boards to devicetree implemenation, we should keep information about external circuitry where they belong - the individual drivers.
This patch adds a way to specify a GPIO to drive the (optional) external pull-up logic, rather than using a function pointer for that.

I do not object against this patch, but I have rather limited knowledge about gpio driver. Patch looks good, feel free to add my

Acked-by Evgeniy Polyakov <[email protected]>

Greg, please pull it into your tree, if Ville does not object. Thank you.

Signature · 1 line

	Evgeniy Polyakov

Daniel Mack

author

2012-07-30 17:35 UTC #4 lore raw

Quoted 9 lines · Evgeniy Polyakov

Hi
Sorry for long delay
Quoted 4 lines · Daniel Mack
In the process of porting boards to devicetree implemenation, we should keep information about external circuitry where they belong - the individual drivers.

Quoted 5 lines

Quoted 2 lines
This patch adds a way to specify a GPIO to drive the (optional) external pull-up logic, rather than using a function pointer for that.
I do not object against this patch, but I have rather limited knowledge about gpio driver.

The first patch just offers a new way to drive an optional external pull-up and as nobody uses that feature currently, there can't be any breakage.

And the second patch doesn't change any functional behaviour of the driver, it simply offers a new way to instanciate it.

Quoted 2 lines

Patch looks good, feel free to add my
Acked-by Evgeniy Polyakov <[email protected]>

Thanks!

Quoted 4 lines

Greg, please pull it into your tree, if Ville does not object. Thank you.

Evgeniy Polyakov

Acked-by on v1 1/2

2012-07-30 18:12 UTC #5 lore raw

In reply to Daniel Mack

Sorry for long delay
[resend with fixed Greg's address :)]

Quoted 6 lines · Daniel Mack

In the process of porting boards to devicetree implemenation, we should keep information about external circuitry where they belong - the individual drivers.
This patch adds a way to specify a GPIO to drive the (optional) external pull-up logic, rather than using a function pointer for that.

I do not object against this patch, but I have rather limited knowledge about gpio driver. Patch looks good, feel free to add my

Acked-by Evgeniy Polyakov <[email protected]>

Greg, please pull it into your tree, if Ville does not object. Thank you.

Signature · 1 line

	Evgeniy Polyakov

Ville Syrjälä

2012-07-31 01:07 UTC #6 lore raw

Quoted 10 lines · Evgeniy Polyakov

Hi
Sorry for long delay
[resend with fixed Greg's address :)]
Quoted 4 lines · Daniel Mack
In the process of porting boards to devicetree implemenation, we should keep information about external circuitry where they belong - the individual drivers.

Quoted 8 lines

Quoted 2 lines
This patch adds a way to specify a GPIO to drive the (optional) external pull-up logic, rather than using a function pointer for that.
I do not object against this patch, but I have rather limited knowledge about gpio driver. Patch looks good, feel free to add my
Acked-by Evgeniy Polyakov <[email protected]>
Greg, please pull it into your tree, if Ville does not object.

No objections from me.

The only cosmetic issue I spotted was 'pdata' being assigned twice in w1_gpio_probe().

Signature · 3 lines

Ville Syrjälä
[email protected]
http://www.sci.fi/~syrjala/

Daniel Mack

author Acked-by Evgeniy Polyakov Acked-by Ville Syrjala

2012-07-31 06:19 UTC #7 lore raw

Quoted 14 lines · Ville Syrjälä

Quoted 13 lines · Evgeniy Polyakov
Hi
Sorry for long delay
[resend with fixed Greg's address :)]
Quoted 6 lines · Daniel Mack
In the process of porting boards to devicetree implemenation, we should keep information about external circuitry where they belong - the individual drivers.
This patch adds a way to specify a GPIO to drive the (optional) external pull-up logic, rather than using a function pointer for that.

Quoted 7 lines

Quoted 5 lines
I do not object against this patch, but I have rather limited knowledge about gpio driver. Patch looks good, feel free to add my
Acked-by Evgeniy Polyakov <[email protected]>
Greg, please pull it into your tree, if Ville does not object.
No objections from me.

I take this as an Acked-by:?

Quoted 2 lines

The only cosmetic issue I spotted was 'pdata' being assigned twice in w1_gpio_probe().

Right. Amended patch below.

Thanks a lot,
Daniel

Quoted 1 line

From 37385d15bbd9dbc637f1e46d962c90be5eeb1409 Mon Sep 17 00:00:00 2001

From: Daniel Mack <[email protected]>
Date: Mon, 23 Jul 2012 16:36:35 +0200
Subject: [PATCH v2] onewire: w1-gpio: add DT bindings

This patch add DT bindings to the w1-gpio driver, along with some documentation on how to use them.

Signed-off-by Daniel Mack <[email protected]>

Acked-by Evgeniy Polyakov <[email protected]>

Acked-by Ville Syrjala <[email protected]>

---
 Documentation/devicetree/bindings/w1/w1-gpio.txt | 22 +++++++++++
 drivers/w1/masters/w1-gpio.c                     | 48 +++++++++++++++++++++++-
 2 files changed, 69 insertions(+), 1 deletion(-)
 create mode 100644 Documentation/devicetree/bindings/w1/w1-gpio.txt

Documentation/devicetree/bindings/w1/w1-gpio.txt new file +22 -0

new file mode 100644
index 0000000..6e09c35
@@ -0,0 +1,22 @@
+w1-gpio devicetree bindings
+
+Required properties:
+
+ - compatible: "w1-gpio"
+ - gpios: one or two GPIO specs:
+		- the first one is used as data I/O pin
+		- the second one is optional. If specified, it is used as
+		  enable pin for an external pin pullup.
+
+Optional properties:
+
+ - linux,open-drain: if specified, the data pin is considered in
+		     open-drain mode.
+
+Examples:
+
+	onewire@0 {
+		compatible = "w1-gpio";
+		gpios = <&gpio 126 0>, <&gpio 105 0>;
+	};
+

drivers/w1/masters/w1-gpio.c +47 -1

71 lines not on this page · raw message

Signature · 1 line

1.7.11.2

Ville Syrjälä

Acked-by

2012-08-01 16:57 UTC #8 lore raw

Quoted 19 lines · Daniel Mack

Quoted 18 lines · Ville Syrjälä
Quoted 17 lines · Evgeniy Polyakov
Hi
Sorry for long delay
[resend with fixed Greg's address :)]
Quoted 6 lines · Daniel Mack
In the process of porting boards to devicetree implemenation, we should keep information about external circuitry where they belong - the individual drivers.
This patch adds a way to specify a GPIO to drive the (optional) external pull-up logic, rather than using a function pointer for that.
I do not object against this patch, but I have rather limited knowledge about gpio driver. Patch looks good, feel free to add my
Acked-by Evgeniy Polyakov <[email protected]>

Quoted 5 lines

Quoted 3 lines
Quoted 1 line
Greg, please pull it into your tree, if Ville does not object.
No objections from me.
I take this as an Acked-by:?

Sure, why not :)

Acked-by Ville Syrjälä <[email protected]>

Signature · 3 lines

Ville Syrjälä
[email protected]
http://www.sci.fi/~syrjala/

Daniel Mack

author

2012-08-05 16:18 UTC #9 lore raw

Quoted 22 lines · Ville Syrjälä

Quoted 21 lines · Daniel Mack
Quoted 20 lines · Ville Syrjälä
Quoted 18 lines · Evgeniy Polyakov
Hi
Sorry for long delay
[resend with fixed Greg's address :)]
Quoted 6 lines · Daniel Mack
In the process of porting boards to devicetree implemenation, we should keep information about external circuitry where they belong - the individual drivers.
This patch adds a way to specify a GPIO to drive the (optional) external pull-up logic, rather than using a function pointer for that.
I do not object against this patch, but I have rather limited knowledge about gpio driver. Patch looks good, feel free to add my
Acked-by Evgeniy Polyakov <[email protected]>
Greg, please pull it into your tree, if Ville does not object.

Quoted 8 lines

Quoted 3 lines
Quoted 1 line
No objections from me.
I take this as an Acked-by:?
Sure, why not :)
Acked-by Ville Syrjälä <[email protected]>

Greg, could you queue up these two patches somewhere?

Thanks,
Daniel

GregKH

2012-08-16 16:57 UTC #10 lore raw

Quoted 25 lines · Daniel Mack

Quoted 24 lines · Ville Syrjälä
Quoted 23 lines · Daniel Mack
Quoted 21 lines · Ville Syrjälä
Quoted 10 lines · Evgeniy Polyakov
Hi
Sorry for long delay
[resend with fixed Greg's address :)]
Quoted 4 lines · Daniel Mack
In the process of porting boards to devicetree implemenation, we should keep information about external circuitry where they belong - the individual drivers.
Quoted 8 lines
Quoted 2 lines
This patch adds a way to specify a GPIO to drive the (optional) external pull-up logic, rather than using a function pointer for that.
I do not object against this patch, but I have rather limited knowledge about gpio driver. Patch looks good, feel free to add my
Acked-by Evgeniy Polyakov <[email protected]>
Greg, please pull it into your tree, if Ville does not object.
No objections from me.

Quoted 8 lines

Quoted 6 lines
Quoted 1 line
I take this as an Acked-by:?
Sure, why not :)
Acked-by Ville Syrjälä <[email protected]>
Greg, could you queue up these two patches somewhere?

Yes, I now have them, thanks.

greg k-h