Skip to contentSkip to conversation

at24: move write-protect pin handling to nvmem core

PATCH Applied Khouloud Touil 7 participants 18 messages Jan 7, 2020 – Feb 17, 2020

Subsystems (5)
lore
Share & tools
lore
#0 lore raw

v4 cover at24: move write-protect pin handling to nvmem core

The write-protect pin handling looks like a standard property that could benefit other users if available in the core nvmem framework.

Instead of modifying all the drivers to check this pin, make the nvmem subsystem check if the write-protect GPIO being passed through the nvmem_config or defined in the device tree and pull it low whenever writing to the memory.

This patchset:

- adds support for the write-protect pin split into two parts. The first patch modifies modifies the relevant binding document, while the second modifies the nvmem code to pull the write-protect GPIO low (if present) during write operations.

- removes support for the write-protect pin split into two parts. The first patch modifies the relevant binding document to make the wp-gpio a reference to the property defined by nvmem , while the second removes the relevant code in the at24 driver.

- adds reference in the at25 binding document for the wp-gpios property as it uses nvmem subsystem.

Changes since v1:
-Add an explenation on how the wp-gpios works -Keep reference to the wp-gpios in the at24 binding

Changes since v2:
-Use the flag GPIO_ACTIVE_HIGH instead of 0

Changes since v3:
-Keep the example of the wp-gpios in the at25 bindings -Add reference for the wp-gpios property in the at25 binding

Khouloud Touil (5):
dt-bindings: nvmem: new optional property wp-gpios
nvmem: add support for the write-protect pin
dt-bindings: at24: make wp-gpios a reference to the property defined
by nvmem
dt-bindings: at25: add reference for the wp-gpios property
eeprom: at24: remove the write-protect pin support

 .../devicetree/bindings/eeprom/at24.yaml      |  5 +----
 .../devicetree/bindings/eeprom/at25.txt       |  2 ++
 .../devicetree/bindings/nvmem/nvmem.yaml      | 11 +++++++++++
 drivers/misc/eeprom/at24.c                    |  9 ---------
 drivers/nvmem/core.c                          | 19 +++++++++++++++++--
 drivers/nvmem/nvmem.h                         |  2 ++
 include/linux/nvmem-provider.h                |  3 +++
 7 files changed, 36 insertions(+), 15 deletions(-)
Signature · 1 line
2.17.1
#1 lore raw

PATCH v4 1/5 dt-bindings: nvmem: new optional property wp-gpios

Several memories have a write-protect pin, that when pulled high, it blocks the write operation.

On some boards, this pin is connected to a GPIO and pulled high by default, which forces the user to manually change its state before writing.

Instead of modifying all the memory drivers to check this pin, make the NVMEM subsystem check if the write-protect GPIO being passed through the nvmem_config or defined in the device tree and pull it low whenever writing to the memory.

Add a new optional property to the device tree binding document, which allows to specify the GPIO line to which the write-protect pin is connected.

---
 Documentation/devicetree/bindings/nvmem/nvmem.yaml | 11 +++++++++++
 1 file changed, 11 insertions(+)
Documentation/devicetree/bindings/nvmem/nvmem.yaml +11 -0
index 1c75a059206c..b43c6c65294e 100644
@@ -34,6 +34,14 @@ properties:
     description:
       Mark the provider as read only.
 
+  wp-gpios:
+    description:
+      GPIO to which the write-protect pin of the chip is connected.
+      The write-protect GPIO is asserted, when it's driven high
+      (logical '1') to block the write operation. It's deasserted,
+      when it's driven low (logical '0') to allow writing.
+    maxItems: 1
+
 patternProperties:
   "^.*@[0-9a-f]+$":
     type: object
@@ -63,9 +71,12 @@ patternProperties:
 
 examples:
   - |
+      #include <dt-bindings/gpio/gpio.h>
+
       qfprom: eeprom@700000 {
           #address-cells = <1>;
           #size-cells = <1>;
+          wp-gpios = <&gpio1 3 GPIO_ACTIVE_HIGH>;
 
           /* ... */
 
Signature · 1 line
2.17.1
Reviewed-by on v4 1/5
#2 lore raw

Re: [PATCH v4 1/5] dt-bindings: nvmem: new optional property wp-gpios

Quoted 12 lines · Khouloud Touil

Several memories have a write-protect pin, that when pulled high, it blocks the write operation.

On some boards, this pin is connected to a GPIO and pulled high by default, which forces the user to manually change its state before writing.

Instead of modifying all the memory drivers to check this pin, make the NVMEM subsystem check if the write-protect GPIO being passed through the nvmem_config or defined in the device tree and pull it low whenever writing to the memory.

Quoted 5 lines

Add a new optional property to the device tree binding document, which allows to specify the GPIO line to which the write-protect pin is connected.

Yours,
Linus Walleij

maintainer Reviewed-by on v4 1/5
#3 lore raw

Re: [PATCH v4 1/5] dt-bindings: nvmem: new optional property wp-gpios

In reply to Khouloud Touil
Quoted 12 lines · Khouloud Touil

Several memories have a write-protect pin, that when pulled high, it blocks the write operation.

On some boards, this pin is connected to a GPIO and pulled high by default, which forces the user to manually change its state before writing.

Instead of modifying all the memory drivers to check this pin, make the NVMEM subsystem check if the write-protect GPIO being passed through the nvmem_config or defined in the device tree and pull it low whenever writing to the memory.

Quoted 9 lines

Add a new optional property to the device tree binding document, which allows to specify the GPIO line to which the write-protect pin is connected.

---
 Documentation/devicetree/bindings/nvmem/nvmem.yaml | 11 +++++++++++
 1 file changed, 11 insertions(+)
#4 lore raw

PATCH v4 2/5 nvmem: add support for the write-protect pin

In reply to Khouloud Touil

The write-protect pin handling looks like a standard property that could benefit other users if available in the core nvmem framework.

Instead of modifying all the memory drivers to check this pin, make the NVMEM subsystem check if the write-protect GPIO being passed through the nvmem_config or defined in the device tree and pull it low whenever writing to the memory.

There was a suggestion for introducing the gpiodesc from pdata, but as pdata is already removed it could be replaced by adding it to nvmem_config.

Reference: https://lists.96boards.org/pipermail/dev/2018-August/001056.html

---
 drivers/nvmem/core.c           | 19 +++++++++++++++++--
 drivers/nvmem/nvmem.h          |  2 ++
 include/linux/nvmem-provider.h |  3 +++
 3 files changed, 22 insertions(+), 2 deletions(-)
drivers/nvmem/core.c +17 -2
index 9f1ee9c766ec..3e1c94c4eee8 100644
@@ -15,6 +15,7 @@
 #include <linux/module.h>
 #include <linux/nvmem-consumer.h>
 #include <linux/nvmem-provider.h>
+#include <linux/gpio/consumer.h>
 #include <linux/of.h>
 #include <linux/slab.h>
 #include "nvmem.h"
@@ -54,8 +55,14 @@ static int nvmem_reg_read(struct nvmem_device *nvmem, unsigned int offset,
 static int nvmem_reg_write(struct nvmem_device *nvmem, unsigned int offset,
 			   void *val, size_t bytes)
 {
-	if (nvmem->reg_write)
-		return nvmem->reg_write(nvmem->priv, offset, val, bytes);
+	int ret;
+
+	if (nvmem->reg_write) {
+		gpiod_set_value_cansleep(nvmem->wp_gpio, 0);
+		ret = nvmem->reg_write(nvmem->priv, offset, val, bytes);
+		gpiod_set_value_cansleep(nvmem->wp_gpio, 1);
+		return ret;
+	}
 
 	return -EINVAL;
 }
@@ -338,6 +345,14 @@ struct nvmem_device *nvmem_register(const struct nvmem_config *config)
 		kfree(nvmem);
 		return ERR_PTR(rval);
 	}
+	if (config->wp_gpio)
+		nvmem->wp_gpio = config->wp_gpio;
+	else
+		nvmem->wp_gpio = gpiod_get_optional(config->dev, "wp",
+						    GPIOD_OUT_HIGH);
+	if (IS_ERR(nvmem->wp_gpio))
+		return PTR_ERR(nvmem->wp_gpio);
+
 
 	kref_init(&nvmem->refcnt);
 	INIT_LIST_HEAD(&nvmem->cells);
drivers/nvmem/nvmem.h +2 -0
index eb8ed7121fa3..be0d66d75c8a 100644
@@ -9,6 +9,7 @@
 #include <linux/list.h>
 #include <linux/nvmem-consumer.h>
 #include <linux/nvmem-provider.h>
+#include <linux/gpio/consumer.h>
 
 struct nvmem_device {
 	struct module		*owner;
@@ -26,6 +27,7 @@ struct nvmem_device {
 	struct list_head	cells;
 	nvmem_reg_read_t	reg_read;
 	nvmem_reg_write_t	reg_write;
+	struct gpio_desc	*wp_gpio;
 	void *priv;
 };
 
include/linux/nvmem-provider.h +3 -0
index fe051323be0a..6d6f8e5d24c9 100644
@@ -11,6 +11,7 @@
 
 #include <linux/err.h>
 #include <linux/errno.h>
+#include <linux/gpio/consumer.h>
 
 struct nvmem_device;
 struct nvmem_cell_info;
@@ -45,6 +46,7 @@ enum nvmem_type {
  * @word_size:	Minimum read/write access granularity.
  * @stride:	Minimum read/write access stride.
  * @priv:	User context passed to read/write callbacks.
+ * @wp-gpio:   Write protect pin
  *
  * Note: A default "nvmem<id>" name will be assigned to the device if
  * no name is specified in its configuration. In such case "<id>" is
@@ -58,6 +60,7 @@ struct nvmem_config {
 	const char		*name;
 	int			id;
 	struct module		*owner;
+	struct gpio_desc	*wp_gpio;
 	const struct nvmem_cell_info	*cells;
 	int			ncells;
 	enum nvmem_type		type;
Signature · 1 line
2.17.1
#5 lore raw

Re: [PATCH v4 2/5] nvmem: add support for the write-protect pin

Hi Khouloud,

Quoted 12 lines · Khouloud Touil

The write-protect pin handling looks like a standard property that could benefit other users if available in the core nvmem framework.

Instead of modifying all the memory drivers to check this pin, make the NVMEM subsystem check if the write-protect GPIO being passed through the nvmem_config or defined in the device tree and pull it low whenever writing to the memory.

There was a suggestion for introducing the gpiodesc from pdata, but as pdata is already removed it could be replaced by adding it to nvmem_config.

Quoted 5 lines

Reference: https://lists.96boards.org/pipermail/dev/2018-August/001056.html

Thanks for your patch!

Quoted 28 lines
drivers/nvmem/core.c +9 -2
@@ -15,6 +15,7 @@
 #include <linux/module.h>
 #include <linux/nvmem-consumer.h>
 #include <linux/nvmem-provider.h>
+#include <linux/gpio/consumer.h>
 #include <linux/of.h>
 #include <linux/slab.h>
 #include "nvmem.h"
@@ -54,8 +55,14 @@ static int nvmem_reg_read(struct nvmem_device *nvmem, unsigned int offset,
 static int nvmem_reg_write(struct nvmem_device *nvmem, unsigned int offset,
                           void *val, size_t bytes)
 {
-       if (nvmem->reg_write)
-               return nvmem->reg_write(nvmem->priv, offset, val, bytes);
+       int ret;
+
+       if (nvmem->reg_write) {
+               gpiod_set_value_cansleep(nvmem->wp_gpio, 0);
+               ret = nvmem->reg_write(nvmem->priv, offset, val, bytes);
+               gpiod_set_value_cansleep(nvmem->wp_gpio, 1);
+               return ret;
+       }
 
        return -EINVAL;
 }

drivers/nvmem/core.c struct nvmem_device *nvmem_register(const struct nvmem_config *config) · line 352

Quoted 8 lines
drivers/nvmem/core.c +5 -0
@@ -338,6 +345,14 @@ struct nvmem_device *nvmem_register(const struct nvmem_config *config)
                kfree(nvmem);
                return ERR_PTR(rval);
        }
+       if (config->wp_gpio)
+               nvmem->wp_gpio = config->wp_gpio;
+       else
+               nvmem->wp_gpio = gpiod_get_optional(config->dev, "wp",
+                                                   GPIOD_OUT_HIGH);

Shouldn't this GPIO be released in nvmem_release(), by calling gpiod_put()?

Once that's implemented, I assume it will be auto-released on registration failure by the call to put_device()?

drivers/nvmem/core.c struct nvmem_device *nvmem_register(const struct nvmem_config *config) · line 358

Quoted 6 lines
drivers/nvmem/core.c +3 -0
+       if (IS_ERR(nvmem->wp_gpio))
+               return PTR_ERR(nvmem->wp_gpio);
+
 
        kref_init(&nvmem->refcnt);
        INIT_LIST_HEAD(&nvmem->cells);

Gr{oetje,eeting}s,

Geert

Signature · 5 lines
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds
#6 lore raw

Re: [PATCH v4 2/5] nvmem: add support for the write-protect pin

Quoted 63 lines · Geert Uytterhoeven

Hi Khouloud,

Quoted 12 lines · Khouloud Touil

The write-protect pin handling looks like a standard property that could benefit other users if available in the core nvmem framework.

Instead of modifying all the memory drivers to check this pin, make the NVMEM subsystem check if the write-protect GPIO being passed through the nvmem_config or defined in the device tree and pull it low whenever writing to the memory.

There was a suggestion for introducing the gpiodesc from pdata, but as pdata is already removed it could be replaced by adding it to nvmem_config.

Quoted 5 lines

Reference: https://lists.96boards.org/pipermail/dev/2018-August/001056.html

Thanks for your patch!

Quoted 28 lines
drivers/nvmem/core.c +9 -2
@@ -15,6 +15,7 @@
 #include <linux/module.h>
 #include <linux/nvmem-consumer.h>
 #include <linux/nvmem-provider.h>
+#include <linux/gpio/consumer.h>
 #include <linux/of.h>
 #include <linux/slab.h>
 #include "nvmem.h"
@@ -54,8 +55,14 @@ static int nvmem_reg_read(struct nvmem_device *nvmem, unsigned int offset,
 static int nvmem_reg_write(struct nvmem_device *nvmem, unsigned int offset,
                           void *val, size_t bytes)
 {
-       if (nvmem->reg_write)
-               return nvmem->reg_write(nvmem->priv, offset, val, bytes);
+       int ret;
+
+       if (nvmem->reg_write) {
+               gpiod_set_value_cansleep(nvmem->wp_gpio, 0);
+               ret = nvmem->reg_write(nvmem->priv, offset, val, bytes);
+               gpiod_set_value_cansleep(nvmem->wp_gpio, 1);
+               return ret;
+       }
 
        return -EINVAL;
 }

drivers/nvmem/core.c struct nvmem_device *nvmem_register(const struct nvmem_config *config) · line 352

Quoted 8 lines
drivers/nvmem/core.c +5 -0
@@ -338,6 +345,14 @@ struct nvmem_device *nvmem_register(const struct nvmem_config *config)
                kfree(nvmem);
                return ERR_PTR(rval);
        }
+       if (config->wp_gpio)
+               nvmem->wp_gpio = config->wp_gpio;
+       else
+               nvmem->wp_gpio = gpiod_get_optional(config->dev, "wp",
+                                                   GPIOD_OUT_HIGH);

Shouldn't this GPIO be released in nvmem_release(), by calling gpiod_put()?

Quoted 2 lines

Once that's implemented, I assume it will be auto-released on registration failure by the call to put_device()?

Hello Geert,

Thanks for your review.
Yes you are right, I will add that

Khouloud,

Quoted 18 lines

drivers/nvmem/core.c struct nvmem_device *nvmem_register(const struct nvmem_config *config) · line 358

Quoted 7 lines
drivers/nvmem/core.c +3 -0
+       if (IS_ERR(nvmem->wp_gpio))
+               return PTR_ERR(nvmem->wp_gpio);
+
 
        kref_init(&nvmem->refcnt);
        INIT_LIST_HEAD(&nvmem->cells);

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

#7 lore raw

Re: [PATCH v4 2/5] nvmem: add support for the write-protect pin

In reply to Geert Uytterhoeven

czw., 30 sty 2020 o 09:06 Geert Uytterhoeven <[email protected]> napisał(a):

Quoted 49 lines

Hi Khouloud,

Quoted 12 lines · Khouloud Touil

The write-protect pin handling looks like a standard property that could benefit other users if available in the core nvmem framework.

Instead of modifying all the memory drivers to check this pin, make the NVMEM subsystem check if the write-protect GPIO being passed through the nvmem_config or defined in the device tree and pull it low whenever writing to the memory.

There was a suggestion for introducing the gpiodesc from pdata, but as pdata is already removed it could be replaced by adding it to nvmem_config.

Quoted 5 lines

Reference: https://lists.96boards.org/pipermail/dev/2018-August/001056.html

Thanks for your patch!

Quoted 25 lines
drivers/nvmem/core.c +9 -2
@@ -15,6 +15,7 @@
 #include <linux/module.h>
 #include <linux/nvmem-consumer.h>
 #include <linux/nvmem-provider.h>
+#include <linux/gpio/consumer.h>
 #include <linux/of.h>
 #include <linux/slab.h>
 #include "nvmem.h"
@@ -54,8 +55,14 @@ static int nvmem_reg_read(struct nvmem_device *nvmem, unsigned int offset,
 static int nvmem_reg_write(struct nvmem_device *nvmem, unsigned int offset,
                           void *val, size_t bytes)
 {
-       if (nvmem->reg_write)
-               return nvmem->reg_write(nvmem->priv, offset, val, bytes);
+       int ret;
+
+       if (nvmem->reg_write) {
+               gpiod_set_value_cansleep(nvmem->wp_gpio, 0);
+               ret = nvmem->reg_write(nvmem->priv, offset, val, bytes);
+               gpiod_set_value_cansleep(nvmem->wp_gpio, 1);
+               return ret;
+       }
 
Quoted 14 lines

drivers/nvmem/core.c struct nvmem_device *nvmem_register(const struct nvmem_config *config) · line 352

Quoted 11 lines
drivers/nvmem/core.c +5 -0
        return -EINVAL;
 }
@@ -338,6 +345,14 @@ struct nvmem_device *nvmem_register(const struct nvmem_config *config)
                kfree(nvmem);
                return ERR_PTR(rval);
        }
+       if (config->wp_gpio)
+               nvmem->wp_gpio = config->wp_gpio;
+       else
+               nvmem->wp_gpio = gpiod_get_optional(config->dev, "wp",
+                                                   GPIOD_OUT_HIGH);

Shouldn't this GPIO be released in nvmem_release(), by calling gpiod_put()?

Hi Geert,

Khouloud already sent out a patch but I think it still doesn't fix all the problems.

While we should call gpiod_put() for the descs we request - we must not do it for the desc we get over the config structure. Unless... we make descs reference counted with kref and add gpiod_ref() helper. That way we could increase the reference counter in the upper branch of the if and not do it in the lower. Calling gpiod_put() would internally call kref_put(). Does it make sense? I think that a function that's called gpiod_put() but doesn't really use reference counting is misleading anyway.

Quoted 2 lines

Once that's implemented, I assume it will be auto-released on registration failure by the call to put_device()?

No, I think this is another leak - why would put_device() lead to freeing any resources? Am I missing something?

Bart

#8 lore raw

Re: [PATCH v4 2/5] nvmem: add support for the write-protect pin

Hi Bartosz,

Quoted 62 lines · Bartosz Golaszewski

czw., 30 sty 2020 o 09:06 Geert Uytterhoeven <[email protected]> napisał(a):

Quoted 46 lines · Khouloud Touil
Quoted 12 lines

The write-protect pin handling looks like a standard property that could benefit other users if available in the core nvmem framework.

Instead of modifying all the memory drivers to check this pin, make the NVMEM subsystem check if the write-protect GPIO being passed through the nvmem_config or defined in the device tree and pull it low whenever writing to the memory.

There was a suggestion for introducing the gpiodesc from pdata, but as pdata is already removed it could be replaced by adding it to nvmem_config.

Quoted 5 lines

Reference: https://lists.96boards.org/pipermail/dev/2018-August/001056.html

Thanks for your patch!

Quoted 25 lines
drivers/nvmem/core.c +9 -2
@@ -15,6 +15,7 @@
 #include <linux/module.h>
 #include <linux/nvmem-consumer.h>
 #include <linux/nvmem-provider.h>
+#include <linux/gpio/consumer.h>
 #include <linux/of.h>
 #include <linux/slab.h>
 #include "nvmem.h"
@@ -54,8 +55,14 @@ static int nvmem_reg_read(struct nvmem_device *nvmem, unsigned int offset,
 static int nvmem_reg_write(struct nvmem_device *nvmem, unsigned int offset,
                           void *val, size_t bytes)
 {
-       if (nvmem->reg_write)
-               return nvmem->reg_write(nvmem->priv, offset, val, bytes);
+       int ret;
+
+       if (nvmem->reg_write) {
+               gpiod_set_value_cansleep(nvmem->wp_gpio, 0);
+               ret = nvmem->reg_write(nvmem->priv, offset, val, bytes);
+               gpiod_set_value_cansleep(nvmem->wp_gpio, 1);
+               return ret;
+       }
 
Quoted 14 lines

drivers/nvmem/core.c struct nvmem_device *nvmem_register(const struct nvmem_config *config) · line 352

Quoted 11 lines
drivers/nvmem/core.c +5 -0
        return -EINVAL;
 }
@@ -338,6 +345,14 @@ struct nvmem_device *nvmem_register(const struct nvmem_config *config)
                kfree(nvmem);
                return ERR_PTR(rval);
        }
+       if (config->wp_gpio)
+               nvmem->wp_gpio = config->wp_gpio;
+       else
+               nvmem->wp_gpio = gpiod_get_optional(config->dev, "wp",
+                                                   GPIOD_OUT_HIGH);

Shouldn't this GPIO be released in nvmem_release(), by calling gpiod_put()?

Quoted 7 lines

Hi Geert,

Khouloud already sent out a patch but I think it still doesn't fix all the problems.

While we should call gpiod_put() for the descs we request - we must not do it for the desc we get over the config structure. Unless... we

That's true.

Quoted 6 lines

make descs reference counted with kref and add gpiod_ref() helper. That way we could increase the reference counter in the upper branch of the if and not do it in the lower. Calling gpiod_put() would internally call kref_put(). Does it make sense? I think that a function that's called gpiod_put() but doesn't really use reference counting is misleading anyway.

Yep.

Quoted 5 lines
Quoted 2 lines

Once that's implemented, I assume it will be auto-released on registration failure by the call to put_device()?

No, I think this is another leak - why would put_device() lead to freeing any resources? Am I missing something?

Sorry, I don't remember why I wrote that part...

Anyway, requested GPIOs should be released on failure, and on unregistration.

Gr{oetje,eeting}s,

Geert

Signature · 5 lines
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
                                -- Linus Torvalds
#9 lore raw

PATCH v4 3/5 dt-bindings: at24: make wp-gpios a reference to the property defined by nvmem

In reply to Khouloud Touil

NVMEM framework is an interface for the at24 EEPROMs as well as for other drivers, instead of passing the wp-gpios over the different drivers each time, it would be better to pass it over the NVMEM subsystem once and for all.

Making wp-gpios a reference to the property defined by nvmem.

---
 Documentation/devicetree/bindings/eeprom/at24.yaml | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)
Documentation/devicetree/bindings/eeprom/at24.yaml +1 -4
index e8778560d966..767959941399 100644
@@ -145,10 +145,7 @@ properties:
       over reads to the next slave address. Please consult the manual of
       your device.
 
-  wp-gpios:
-    description:
-      GPIO to which the write-protect pin of the chip is connected.
-    maxItems: 1
+  wp-gpios: true
 
   address-width:
     allOf:
Signature · 1 line
2.17.1
maintainer Reviewed-by on v4 3/5
#10 lore raw

Re: [PATCH v4 3/5] dt-bindings: at24: make wp-gpios a reference to the property defined by nvmem

Quoted 5 lines · Khouloud Touil

NVMEM framework is an interface for the at24 EEPROMs as well as for other drivers, instead of passing the wp-gpios over the different drivers each time, it would be better to pass it over the NVMEM subsystem once and for all.

Quoted 8 lines

Making wp-gpios a reference to the property defined by nvmem.

---
 Documentation/devicetree/bindings/eeprom/at24.yaml | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)
#11 lore raw

PATCH v4 4/5 dt-bindings: at25: add reference for the wp-gpios property

In reply to Khouloud Touil

As the at25 uses the NVMEM subsystem, and the property is now being handled, adding reference for it in the device tree binding document, which allows to specify the GPIO line to which the write-protect pin is connected.

---
 Documentation/devicetree/bindings/eeprom/at25.txt | 2 ++
 1 file changed, 2 insertions(+)
Documentation/devicetree/bindings/eeprom/at25.txt +2 -0
index 42577dd113dd..fcacd97abd0a 100644
@@ -20,6 +20,7 @@ Optional properties:
 - spi-cpha : SPI shifted clock phase, as per spi-bus bindings.
 - spi-cpol : SPI inverse clock polarity, as per spi-bus bindings.
 - read-only : this parameter-less property disables writes to the eeprom
+- wp-gpios : GPIO to which the write-protect pin of the chip is connected
 
 Obsolete legacy properties can be used in place of "size", "pagesize",
 "address-width", and "read-only":
@@ -36,6 +37,7 @@ Example:
 		spi-max-frequency = <5000000>;
 		spi-cpha;
 		spi-cpol;
+		wp-gpios = <&gpio1 3 0>;
 
 		pagesize = <64>;
 		size = <32768>;
Signature · 1 line
2.17.1
maintainer Reviewed-by on v4 4/5
#12 lore raw

Re: [PATCH v4 4/5] dt-bindings: at25: add reference for the wp-gpios property

Quoted 10 lines · Khouloud Touil

As the at25 uses the NVMEM subsystem, and the property is now being handled, adding reference for it in the device tree binding document, which allows to specify the GPIO line to which the write-protect pin is connected.

---
 Documentation/devicetree/bindings/eeprom/at25.txt | 2 ++
 1 file changed, 2 insertions(+)
#13 lore raw

Re: [PATCH v4 4/5] dt-bindings: at25: add reference for the wp-gpios property

śr., 8 sty 2020 o 21:54 Rob Herring <[email protected]> napisał(a):

Quoted 7 lines · Khouloud Touil
Quoted 5 lines

As the at25 uses the NVMEM subsystem, and the property is now being handled, adding reference for it in the device tree binding document, which allows to specify the GPIO line to which the write-protect pin is connected.

Quoted 7 lines
Quoted 5 lines
---
 Documentation/devicetree/bindings/eeprom/at25.txt | 2 ++
 1 file changed, 2 insertions(+)

Hi Greg,

AT25 patches usually go through the char-misc tree. In this case however, the change depends on the other patches in this series. Can you ack this and I'll take it through the AT24 tree exceptionally?

Best regards,
Bartosz Golaszewski

maintainer Acked-by
#14 lore raw

Re: [PATCH v4 4/5] dt-bindings: at25: add reference for the wp-gpios property

Quoted 14 lines · Bartosz Golaszewski

śr., 8 sty 2020 o 21:54 Rob Herring <[email protected]> napisał(a):

Quoted 13 lines · Khouloud Touil
Quoted 10 lines

As the at25 uses the NVMEM subsystem, and the property is now being handled, adding reference for it in the device tree binding document, which allows to specify the GPIO line to which the write-protect pin is connected.

---
 Documentation/devicetree/bindings/eeprom/at25.txt | 2 ++
 1 file changed, 2 insertions(+)
Quoted 7 lines
Quoted 1 line

Hi Greg,

AT25 patches usually go through the char-misc tree. In this case however, the change depends on the other patches in this series. Can you ack this and I'll take it through the AT24 tree exceptionally?

#15 lore raw

Re: [PATCH v4 4/5] dt-bindings: at25: add reference for the wp-gpios property

wt., 14 sty 2020 o 15:42 Greg KH <[email protected]> napisał(a):

Quoted 18 lines · Bartosz Golaszewski
Quoted 16 lines

śr., 8 sty 2020 o 21:54 Rob Herring <[email protected]> napisał(a):

Quoted 14 lines · Khouloud Touil
Quoted 10 lines

As the at25 uses the NVMEM subsystem, and the property is now being handled, adding reference for it in the device tree binding document, which allows to specify the GPIO line to which the write-protect pin is connected.

---
 Documentation/devicetree/bindings/eeprom/at25.txt | 2 ++
 1 file changed, 2 insertions(+)
Quoted 7 lines
Quoted 5 lines

Hi Greg,

AT25 patches usually go through the char-misc tree. In this case however, the change depends on the other patches in this series. Can you ack this and I'll take it through the AT24 tree exceptionally?

Patch applied with Greg's Ack.

Bart

#16 lore raw

PATCH v4 5/5 eeprom: at24: remove the write-protect pin support

In reply to Khouloud Touil

NVMEM framework is an interface for the at24 EEPROMs as well as for other drivers, instead of passing the wp-gpios over the different drivers each time, it would be better to pass it over the NVMEM subsystem once and for all.

Removing the support for the write-protect pin after adding it to the NVMEM subsystem.

---
 drivers/misc/eeprom/at24.c | 9 ---------
 1 file changed, 9 deletions(-)
drivers/misc/eeprom/at24.c +0 -9
index 0681d5fdd538..8fce49a6d9cd 100644
@@ -22,7 +22,6 @@
 #include <linux/nvmem-provider.h>
 #include <linux/regmap.h>
 #include <linux/pm_runtime.h>
-#include <linux/gpio/consumer.h>
 
 /* Address pointer is 16 bit. */
 #define AT24_FLAG_ADDR16	BIT(7)
@@ -89,8 +88,6 @@ struct at24_data {
 
 	struct nvmem_device *nvmem;
 
-	struct gpio_desc *wp_gpio;
-
 	/*
 	 * Some chips tie up multiple I2C addresses; dummy devices reserve
 	 * them for us, and we'll use them with SMBus calls.
@@ -457,12 +454,10 @@ static int at24_write(void *priv, unsigned int off, void *val, size_t count)
 	 * from this host, but not from other I2C masters.
 	 */
 	mutex_lock(&at24->lock);
-	gpiod_set_value_cansleep(at24->wp_gpio, 0);
 
 	while (count) {
 		ret = at24_regmap_write(at24, buf, off, count);
 		if (ret < 0) {
-			gpiod_set_value_cansleep(at24->wp_gpio, 1);
 			mutex_unlock(&at24->lock);
 			pm_runtime_put(dev);
 			return ret;
@@ -472,7 +467,6 @@ static int at24_write(void *priv, unsigned int off, void *val, size_t count)
 		count -= ret;
 	}
 
-	gpiod_set_value_cansleep(at24->wp_gpio, 1);
 	mutex_unlock(&at24->lock);
 
 	pm_runtime_put(dev);
@@ -662,9 +656,6 @@ static int at24_probe(struct i2c_client *client)
 	at24->client[0].client = client;
 	at24->client[0].regmap = regmap;
 
-	at24->wp_gpio = devm_gpiod_get_optional(dev, "wp", GPIOD_OUT_HIGH);
-	if (IS_ERR(at24->wp_gpio))
-		return PTR_ERR(at24->wp_gpio);
 
 	writable = !(flags & AT24_FLAG_READONLY);
 	if (writable) {
Signature · 1 line
2.17.1
#17 lore raw
In reply to Khouloud Touil

wt., 7 sty 2020 o 10:29 Khouloud Touil <[email protected]> napisał(a):

Quoted 32 lines

The write-protect pin handling looks like a standard property that could benefit other users if available in the core nvmem framework.

Instead of modifying all the drivers to check this pin, make the nvmem subsystem check if the write-protect GPIO being passed through the nvmem_config or defined in the device tree and pull it low whenever writing to the memory.

This patchset:

- adds support for the write-protect pin split into two parts. The first patch modifies modifies the relevant binding document, while the second modifies the nvmem code to pull the write-protect GPIO low (if present) during write operations.

- removes support for the write-protect pin split into two parts. The first patch modifies the relevant binding document to make the wp-gpio a reference to the property defined by nvmem , while the second removes the relevant code in the at24 driver.

- adds reference in the at25 binding document for the wp-gpios property as it uses nvmem subsystem.

Changes since v1:

(quoted diff) +0 -2
-Add an explenation on how the wp-gpios works
-Keep reference to the wp-gpios in the at24 binding

Changes since v2:
-Use the flag GPIO_ACTIVE_HIGH instead of 0

Changes since v3:

Review

Quoted 2 lines
(quoted diff) +0 -2
-Keep the example of the wp-gpios in the at25 bindings
-Add reference for the wp-gpios property in the at25 binding

I picked up patches 1-3 & 5 into the at24 tree.

Patch 4 will need an Ack from Greg.

Bart