Skip to contentSkip to conversation

watchdog: Add DA906x PMIC watchdog driver.

PATCH Superseded Markus Pargmann 2 participants 6 messages Aug 13, 2014 – Aug 15, 2014

Subsystems (2)

v3 of 7 Superseded Latest is v7 →

lore
Share & tools
lore
#0 lore raw

PATCH v3 1/1 watchdog: Add DA906x PMIC watchdog driver.

From: Krystian Garbaciak <[email protected]>

This driver supports the watchdog device inside the DA906x PMIC.

---
 drivers/watchdog/Kconfig      |   7 ++
 drivers/watchdog/Makefile     |   1 +
 drivers/watchdog/da9063_wdt.c | 228 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 236 insertions(+)
 create mode 100644 drivers/watchdog/da9063_wdt.c
drivers/watchdog/Kconfig +7 -0
index c845527b503a..8d5c9b33552a 100644
@@ -87,6 +87,13 @@ config DA9055_WATCHDOG
 	  This driver can also be built as a module.  If so, the module
 	  will be called da9055_wdt.
 
+config DA9063_WATCHDOG
+	tristate "Dialog DA9063 Watchdog"
+	depends on MFD_DA9063
+	select WATCHDOG_CORE
+	help
+	  Support for the watchdog in the DA9063 PMIC.
+
 config GPIO_WATCHDOG
 	tristate "Watchdog device controlled through GPIO-line"
 	depends on OF_GPIO
drivers/watchdog/Makefile +1 -0
index 7b8a91ed20e7..ce4a7632d863 100644
@@ -172,6 +172,7 @@ obj-$(CONFIG_XEN_WDT) += xen_wdt.o
 # Architecture Independent
 obj-$(CONFIG_DA9052_WATCHDOG) += da9052_wdt.o
 obj-$(CONFIG_DA9055_WATCHDOG) += da9055_wdt.o
+obj-$(CONFIG_DA9063_WATCHDOG) += da9063_wdt.o
 obj-$(CONFIG_GPIO_WATCHDOG)	+= gpio_wdt.o
 obj-$(CONFIG_WM831X_WATCHDOG) += wm831x_wdt.o
 obj-$(CONFIG_WM8350_WATCHDOG) += wm8350_wdt.o
drivers/watchdog/da9063_wdt.c new file +228 -0

228 lines not on this page · raw message

Signature · 1 line
2.0.1
#1 lore raw

PATCH v3 1/1 watchdog: Add DA906x PMIC watchdog driver.

Hi,

Quoted 7 lines · Markus Pargmann

From: Krystian Garbaciak <[email protected]>

This driver supports the watchdog device inside the DA906x PMIC.

Quoted 6 lines
---
 drivers/watchdog/Kconfig      |   7 ++
 drivers/watchdog/Makefile     |   1 +
 drivers/watchdog/da9063_wdt.c | 228 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 236 insertions(+)
 create mode 100644 drivers/watchdog/da9063_wdt.c

Forgot to add --notes to send-email.

Here are the changes in v3:
- Cleanup error handling for timeout selection and setting
- Fix race condition due to late wdt drvdata setup
- Remove static struct watchdog_device. It is not initialized in the probe function
- Use WATCHDOG_NOWAYOUT_INIT_STATUS for the status
- Remove 0 shift using DA9063_TWDSCALE_SHIFT

Best regards,

Markus Pargmann

Signature · 11 lines
Pengutronix e.K.                           |                             |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20140813/f4c51265/attachment.sig>
maintainer
#2 lore raw

PATCH v3 1/1 watchdog: Add DA906x PMIC watchdog driver.

In reply to Markus Pargmann
Quoted 7 lines · Markus Pargmann

From: Krystian Garbaciak <[email protected]>

This driver supports the watchdog device inside the DA906x PMIC.

Couple of additional comments.

Guenter

Quoted 112 lines
---
  drivers/watchdog/Kconfig      |   7 ++
  drivers/watchdog/Makefile     |   1 +
  drivers/watchdog/da9063_wdt.c | 228 ++++++++++++++++++++++++++++++++++++++++++
  3 files changed, 236 insertions(+)

create mode 100644 drivers/watchdog/da9063_wdt.c

drivers/watchdog/Kconfig +7 -0
index c845527b503a..8d5c9b33552a 100644
@@ -87,6 +87,13 @@ config DA9055_WATCHDOG
  	  This driver can also be built as a module.  If so, the module
  	  will be called da9055_wdt.
 
+config DA9063_WATCHDOG
+	tristate "Dialog DA9063 Watchdog"
+	depends on MFD_DA9063
+	select WATCHDOG_CORE
+	help
+	  Support for the watchdog in the DA9063 PMIC.
+
  config GPIO_WATCHDOG
  	tristate "Watchdog device controlled through GPIO-line"
  	depends on OF_GPIO
drivers/watchdog/Makefile +1 -0
index 7b8a91ed20e7..ce4a7632d863 100644
@@ -172,6 +172,7 @@ obj-$(CONFIG_XEN_WDT) += xen_wdt.o
  # Architecture Independent
  obj-$(CONFIG_DA9052_WATCHDOG) += da9052_wdt.o
  obj-$(CONFIG_DA9055_WATCHDOG) += da9055_wdt.o
+obj-$(CONFIG_DA9063_WATCHDOG) += da9063_wdt.o
  obj-$(CONFIG_GPIO_WATCHDOG)	+= gpio_wdt.o
  obj-$(CONFIG_WM831X_WATCHDOG) += wm831x_wdt.o
  obj-$(CONFIG_WM8350_WATCHDOG) += wm8350_wdt.o
drivers/watchdog/da9063_wdt.c new file +69 -0
new file mode 100644
index 000000000000..2b5ce564c4c1
@@ -0,0 +1,228 @@
+/*
+ * Watchdog driver for DA9063 PMICs.
+ *
+ * Copyright(c) 2012 Dialog Semiconductor Ltd.
+ *
+ * Author: Mariusz Wojtasik <[email protected]>
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/watchdog.h>
+#include <linux/platform_device.h>
+#include <linux/uaccess.h>
+#include <linux/slab.h>
+#include <linux/delay.h>
+#include <linux/mfd/da9063/registers.h>
+#include <linux/mfd/da9063/core.h>
+#include <linux/regmap.h>
+
+/*
+ * Watchdog selector to timeout in seconds.
+ *   0: WDT disabled;
+ *   others: timeout = 2048 ms * 2^(TWDSCALE-1).
+ */
+static const int wdt_timeout[] = { 0, 2, 4, 8, 16, 32, 65, 131 };
+#define DA9063_TWDSCALE_DISABLE		0
+#define DA9063_TWDSCALE_MIN		1
+#define DA9063_TWDSCALE_MAX		(ARRAY_SIZE(wdt_timeout) - 1)
+#define DA9063_WDT_MIN_TIMEOUT		wdt_timeout[DA9063_TWDSCALE_MIN]
+#define DA9063_WDT_MAX_TIMEOUT		wdt_timeout[DA9063_TWDSCALE_MAX]
+#define DA9063_WDG_TIMEOUT		wdt_timeout[3]
+
+struct da9063_watchdog {
+	struct da9063 *da9063;
+	struct watchdog_device wdtdev;
+	struct mutex lock;
+};
+
+static int da9063_wdt_timeout_to_sel(int secs)
+{
+	int i;
+
+	for (i = DA9063_TWDSCALE_MIN; i <= DA9063_TWDSCALE_MAX; i++) {
+		if (wdt_timeout[i] >= secs)
+			return i;
+	}
+
+	return -EINVAL;
+}
+
+static int da9063_wdt_disable(struct da9063 *da9063)
+{
+	return regmap_update_bits(da9063->regmap, DA9063_REG_CONTROL_D,
+			DA9063_TWDSCALE_MASK,
+			DA9063_TWDSCALE_DISABLE);
+}
+
+static int _da9063_wdt_set_timeout(struct da9063 *da9063, unsigned int regval)
+{
+	return regmap_update_bits(da9063->regmap, DA9063_REG_CONTROL_D,
+			DA9063_TWDSCALE_MASK, regval);
+}
+
+static int _da9063_wdt_kick(struct da9063 *da9063)

drivers/watchdog/da9063_wdt.c line 79

Quoted 10 lines
drivers/watchdog/da9063_wdt.c new file +10 -0
+{
+	return regmap_write(da9063->regmap, DA9063_REG_CONTROL_F,
+			DA9063_WATCHDOG);
+}
+
+static int da9063_wdt_start(struct watchdog_device *wdd)
+{
+	struct da9063_watchdog *wdt = watchdog_get_drvdata(wdd);
+	unsigned int selector;
+	int ret = 0;

Unnecessary initialization.

drivers/watchdog/da9063_wdt.c line 86

Quoted 7 lines
drivers/watchdog/da9063_wdt.c new file +7 -0
+
+	mutex_lock(&wdt->lock);
+	selector = da9063_wdt_timeout_to_sel(wdt->wdtdev.timeout);
+	if (selector < 0) {
+		dev_err(wdt->da9063->dev, "Timeout out of range (timeout=%d)\n",
+				wdt->wdtdev.timeout);
+		return selector;

That only works if you move the lock further down, which should be ok. Otherwise you'll need to assign the return code to ret and jump to the unlock below.

Quoted 6 lines
drivers/watchdog/da9063_wdt.c new file +6 -0
+	}
+
+	ret = _da9063_wdt_set_timeout(wdt->da9063, selector);
+	if (ret) {
+		dev_err(wdt->da9063->dev, "Watchdog failed to start (err = %d)\n",
+				ret);

drivers/watchdog/da9063_wdt.c line 102

Quoted 10 lines
drivers/watchdog/da9063_wdt.c new file +10 -0
+	}
+	mutex_unlock(&wdt->lock);
+
+	return ret;
+}
+
+static int da9063_wdt_stop(struct watchdog_device *wdd)
+{
+	struct da9063_watchdog *wdt = watchdog_get_drvdata(wdd);
+	int ret = 0;

Unnecessary initialization.

Quoted 6 lines
drivers/watchdog/da9063_wdt.c new file +6 -0
+
+	mutex_lock(&wdt->lock);
+	ret = da9063_wdt_disable(wdt->da9063);
+	if (ret)
+		dev_alert(wdt->da9063->dev,
+			  "Watchdog failed to stop (err = %d)\n", ret);

drivers/watchdog/da9063_wdt.c line 118

Quoted 10 lines
drivers/watchdog/da9063_wdt.c new file +10 -0
+	mutex_unlock(&wdt->lock);
+
+	return ret;
+}
+
+static int da9063_wdt_ping(struct watchdog_device *wdd)
+{
+	struct da9063_watchdog *wdt = watchdog_get_drvdata(wdd);
+	int ret = 0;
+

Unnecessary initialization.

drivers/watchdog/da9063_wdt.c line 130

Quoted 12 lines
drivers/watchdog/da9063_wdt.c new file +12 -0
+	mutex_lock(&wdt->lock);
+	ret = _da9063_wdt_kick(wdt->da9063);
+	if (ret)
+		dev_alert(wdt->da9063->dev, "Failed to ping the watchdog (err = %d)\n",
+				ret);
+	mutex_unlock(&wdt->lock);
+
+	return ret;
+}
+
+static int da9063_wdt_set_timeout(struct watchdog_device *wdd,
+							unsigned int timeout)

Sometimes it seems to me people go through great pains not to follow Linux indentation rules, and use the "as many different indentations as possible" rule instead ;-). No, I am not enforcing it; that would be up to Wim if he cares.

Quoted 36 lines
drivers/watchdog/da9063_wdt.c new file +36 -0
+{
+	struct da9063_watchdog *wdt = watchdog_get_drvdata(wdd);
+	unsigned int selector;
+	int ret;
+
+	selector = da9063_wdt_timeout_to_sel(timeout);
+	if (selector < 0) {
+		dev_err(wdt->da9063->dev, "Unsupported watchdog timeout, should be between 2 and 131\n");
+		return -EINVAL;
+	}
+
+	mutex_lock(&wdt->lock);
+
+	ret = _da9063_wdt_set_timeout(wdt->da9063, selector);
+	if (ret)
+		dev_err(wdt->da9063->dev, "Failed to set watchdog timeout (err = %d)\n",
+				ret);
+	else
+		wdd->timeout = wdt_timeout[selector];
+
+	mutex_unlock(&wdt->lock);
+
+	return ret;
+}
+
+static const struct watchdog_info da9063_watchdog_info = {
+	.options = WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING,
+	.identity = "DA9063 Watchdog",
+};
+
+static const struct watchdog_ops da9063_watchdog_ops = {
+	.owner = THIS_MODULE,
+	.start = da9063_wdt_start,
+	.stop = da9063_wdt_stop,
+	.ping = da9063_wdt_ping,
+	.set_timeout = da9063_wdt_set_timeout,

drivers/watchdog/da9063_wdt.c line 176

Quoted 10 lines
drivers/watchdog/da9063_wdt.c new file +10 -0
+};
+
+static int da9063_wdt_probe(struct platform_device *pdev)
+{
+	int ret;
+	struct da9063 *da9063 = dev_get_drvdata(pdev->dev.parent);
+	struct da9063_watchdog *wdt;
+
+	if (!da9063)
+		return -EPROBE_DEFER;

Can this ever happen ? Isn't this driver instantiated from its parent, which would ensure that the parent's driver data is set ?

Quoted 6 lines
drivers/watchdog/da9063_wdt.c new file +6 -0
+
+	wdt = devm_kzalloc(&pdev->dev, sizeof(*wdt), GFP_KERNEL);
+	if (!wdt)
+		return -ENOMEM;
+
+	wdt->da9063 = da9063;

drivers/watchdog/da9063_wdt.c line 192

Quoted 10 lines
drivers/watchdog/da9063_wdt.c new file +10 -0
+	mutex_init(&wdt->lock);
+
+	wdt->wdtdev.info = &da9063_watchdog_info;
+	wdt->wdtdev.ops = &da9063_watchdog_ops;
+	wdt->wdtdev.min_timeout = DA9063_WDT_MIN_TIMEOUT;
+	wdt->wdtdev.max_timeout = DA9063_WDT_MAX_TIMEOUT;
+	wdt->wdtdev.timeout = DA9063_WDG_TIMEOUT;
+
+	if (IS_ENABLED(CONFIG_WATCHDOG_NOWAYOUT))
+		wdt->wdtdev.status = WATCHDOG_NOWAYOUT_INIT_STATUS;

You don't need the if() here. That is embedded in WATCHDOG_NOWAYOUT_INIT_STATUS.

Quoted 37 lines
drivers/watchdog/da9063_wdt.c new file +36 -0
+
+	watchdog_set_drvdata(&wdt->wdtdev, wdt);
+	dev_set_drvdata(&pdev->dev, wdt);
+
+	ret = watchdog_register_device(&wdt->wdtdev);
+	if (ret) {
+		dev_err(da9063->dev, "da9063-watchdog registration failed (err = %d)",
+				ret);
+		return ret;
+	}
+
+	return 0;
+}
+
+static int da9063_wdt_remove(struct platform_device *pdev)
+{
+	struct da9063_watchdog *wdt = dev_get_drvdata(&pdev->dev);
+
+	watchdog_unregister_device(&wdt->wdtdev);
+
+	return 0;
+}
+
+static struct platform_driver da9063_wdt_driver = {
+	.probe = da9063_wdt_probe,
+	.remove = da9063_wdt_remove,
+	.driver = {
+		.name = DA9063_DRVNAME_WATCHDOG,
+	},
+};
+module_platform_driver(da9063_wdt_driver);
+
+MODULE_AUTHOR("Mariusz Wojtasik <[email protected]>");
+MODULE_DESCRIPTION("Watchdog driver for Dialog DA9063");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS("platform:" DA9063_DRVNAME_WATCHDOG);
#3 lore raw

PATCH v3 1/1 watchdog: Add DA906x PMIC watchdog driver.

Quoted 129 lines · Guenter Roeck
Quoted 7 lines · Markus Pargmann

From: Krystian Garbaciak <[email protected]>

This driver supports the watchdog device inside the DA906x PMIC.

Couple of additional comments.

Guenter

Quoted 116 lines
---
 drivers/watchdog/Kconfig      |   7 ++
 drivers/watchdog/Makefile     |   1 +
 drivers/watchdog/da9063_wdt.c | 228 ++++++++++++++++++++++++++++++++++++++++++
 3 files changed, 236 insertions(+)
 create mode 100644 drivers/watchdog/da9063_wdt.c
drivers/watchdog/Kconfig +7 -0
index c845527b503a..8d5c9b33552a 100644
@@ -87,6 +87,13 @@ config DA9055_WATCHDOG
 	  This driver can also be built as a module.  If so, the module
 	  will be called da9055_wdt.
 
+config DA9063_WATCHDOG
+	tristate "Dialog DA9063 Watchdog"
+	depends on MFD_DA9063
+	select WATCHDOG_CORE
+	help
+	  Support for the watchdog in the DA9063 PMIC.
+
 config GPIO_WATCHDOG
 	tristate "Watchdog device controlled through GPIO-line"
 	depends on OF_GPIO
drivers/watchdog/Makefile +1 -0
index 7b8a91ed20e7..ce4a7632d863 100644
@@ -172,6 +172,7 @@ obj-$(CONFIG_XEN_WDT) += xen_wdt.o
 # Architecture Independent
 obj-$(CONFIG_DA9052_WATCHDOG) += da9052_wdt.o
 obj-$(CONFIG_DA9055_WATCHDOG) += da9055_wdt.o
+obj-$(CONFIG_DA9063_WATCHDOG) += da9063_wdt.o
 obj-$(CONFIG_GPIO_WATCHDOG)	+= gpio_wdt.o
 obj-$(CONFIG_WM831X_WATCHDOG) += wm831x_wdt.o
 obj-$(CONFIG_WM8350_WATCHDOG) += wm8350_wdt.o
drivers/watchdog/da9063_wdt.c new file +73 -0
new file mode 100644
index 000000000000..2b5ce564c4c1
@@ -0,0 +1,228 @@
+/*
+ * Watchdog driver for DA9063 PMICs.
+ *
+ * Copyright(c) 2012 Dialog Semiconductor Ltd.
+ *
+ * Author: Mariusz Wojtasik <[email protected]>
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/watchdog.h>
+#include <linux/platform_device.h>
+#include <linux/uaccess.h>
+#include <linux/slab.h>
+#include <linux/delay.h>
+#include <linux/mfd/da9063/registers.h>
+#include <linux/mfd/da9063/core.h>
+#include <linux/regmap.h>
+
+/*
+ * Watchdog selector to timeout in seconds.
+ *   0: WDT disabled;
+ *   others: timeout = 2048 ms * 2^(TWDSCALE-1).
+ */
+static const int wdt_timeout[] = { 0, 2, 4, 8, 16, 32, 65, 131 };
+#define DA9063_TWDSCALE_DISABLE		0
+#define DA9063_TWDSCALE_MIN		1
+#define DA9063_TWDSCALE_MAX		(ARRAY_SIZE(wdt_timeout) - 1)
+#define DA9063_WDT_MIN_TIMEOUT		wdt_timeout[DA9063_TWDSCALE_MIN]
+#define DA9063_WDT_MAX_TIMEOUT		wdt_timeout[DA9063_TWDSCALE_MAX]
+#define DA9063_WDG_TIMEOUT		wdt_timeout[3]
+
+struct da9063_watchdog {
+	struct da9063 *da9063;
+	struct watchdog_device wdtdev;
+	struct mutex lock;
+};
+
+static int da9063_wdt_timeout_to_sel(int secs)
+{
+	int i;
+
+	for (i = DA9063_TWDSCALE_MIN; i <= DA9063_TWDSCALE_MAX; i++) {
+		if (wdt_timeout[i] >= secs)
+			return i;
+	}
+
+	return -EINVAL;
+}
+
+static int da9063_wdt_disable(struct da9063 *da9063)
+{
+	return regmap_update_bits(da9063->regmap, DA9063_REG_CONTROL_D,
+			DA9063_TWDSCALE_MASK,
+			DA9063_TWDSCALE_DISABLE);
+}
+
+static int _da9063_wdt_set_timeout(struct da9063 *da9063, unsigned int regval)
+{
+	return regmap_update_bits(da9063->regmap, DA9063_REG_CONTROL_D,
+			DA9063_TWDSCALE_MASK, regval);
+}
+
+static int _da9063_wdt_kick(struct da9063 *da9063)
+{
+	return regmap_write(da9063->regmap, DA9063_REG_CONTROL_F,
+			DA9063_WATCHDOG);
+}
Quoted 8 lines

drivers/watchdog/da9063_wdt.c line 79

Quoted 6 lines
drivers/watchdog/da9063_wdt.c new file +6 -0
+
+static int da9063_wdt_start(struct watchdog_device *wdd)
+{
+	struct da9063_watchdog *wdt = watchdog_get_drvdata(wdd);
+	unsigned int selector;
+	int ret = 0;

Unnecessary initialization.

Thanks, I fixed all these initializations.

Quoted 12 lines

drivers/watchdog/da9063_wdt.c line 86

Quoted 8 lines
drivers/watchdog/da9063_wdt.c new file +7 -0
+
+	mutex_lock(&wdt->lock);
+	selector = da9063_wdt_timeout_to_sel(wdt->wdtdev.timeout);
+	if (selector < 0) {
+		dev_err(wdt->da9063->dev, "Timeout out of range (timeout=%d)\n",
+				wdt->wdtdev.timeout);
+		return selector;

That only works if you move the lock further down, which should be ok. Otherwise you'll need to assign the return code to ret and jump to the unlock below.

Yes, moved.

Quoted 38 lines
Quoted 7 lines
drivers/watchdog/da9063_wdt.c new file +6 -0
+	}
+
+	ret = _da9063_wdt_set_timeout(wdt->da9063, selector);
+	if (ret) {
+		dev_err(wdt->da9063->dev, "Watchdog failed to start (err = %d)\n",
+				ret);

drivers/watchdog/da9063_wdt.c line 102

Quoted 10 lines
drivers/watchdog/da9063_wdt.c new file +10 -0
+	}
+	mutex_unlock(&wdt->lock);
+
+	return ret;
+}
+
+static int da9063_wdt_stop(struct watchdog_device *wdd)
+{
+	struct da9063_watchdog *wdt = watchdog_get_drvdata(wdd);
+	int ret = 0;

Unnecessary initialization.

Quoted 6 lines
drivers/watchdog/da9063_wdt.c new file +6 -0
+
+	mutex_lock(&wdt->lock);
+	ret = da9063_wdt_disable(wdt->da9063);
+	if (ret)
+		dev_alert(wdt->da9063->dev,
+			  "Watchdog failed to stop (err = %d)\n", ret);

drivers/watchdog/da9063_wdt.c line 118

Quoted 10 lines
drivers/watchdog/da9063_wdt.c new file +10 -0
+	mutex_unlock(&wdt->lock);
+
+	return ret;
+}
+
+static int da9063_wdt_ping(struct watchdog_device *wdd)
+{
+	struct da9063_watchdog *wdt = watchdog_get_drvdata(wdd);
+	int ret = 0;
+

Unnecessary initialization.

Quoted 16 lines

drivers/watchdog/da9063_wdt.c line 130

Quoted 12 lines
drivers/watchdog/da9063_wdt.c new file +12 -0
+	mutex_lock(&wdt->lock);
+	ret = _da9063_wdt_kick(wdt->da9063);
+	if (ret)
+		dev_alert(wdt->da9063->dev, "Failed to ping the watchdog (err = %d)\n",
+				ret);
+	mutex_unlock(&wdt->lock);
+
+	return ret;
+}
+
+static int da9063_wdt_set_timeout(struct watchdog_device *wdd,
+							unsigned int timeout)

Sometimes it seems to me people go through great pains not to follow Linux indentation rules, and use the "as many different indentations as possible" rule instead ;-). No, I am not enforcing it; that would be up to Wim if he cares.

I fixed this indentation to the same style as the rest of the driver.

Quoted 42 lines
Quoted 42 lines
drivers/watchdog/da9063_wdt.c new file +41 -0
+{
+	struct da9063_watchdog *wdt = watchdog_get_drvdata(wdd);
+	unsigned int selector;
+	int ret;
+
+	selector = da9063_wdt_timeout_to_sel(timeout);
+	if (selector < 0) {
+		dev_err(wdt->da9063->dev, "Unsupported watchdog timeout, should be between 2 and 131\n");
+		return -EINVAL;
+	}
+
+	mutex_lock(&wdt->lock);
+
+	ret = _da9063_wdt_set_timeout(wdt->da9063, selector);
+	if (ret)
+		dev_err(wdt->da9063->dev, "Failed to set watchdog timeout (err = %d)\n",
+				ret);
+	else
+		wdd->timeout = wdt_timeout[selector];
+
+	mutex_unlock(&wdt->lock);
+
+	return ret;
+}
+
+static const struct watchdog_info da9063_watchdog_info = {
+	.options = WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING,
+	.identity = "DA9063 Watchdog",
+};
+
+static const struct watchdog_ops da9063_watchdog_ops = {
+	.owner = THIS_MODULE,
+	.start = da9063_wdt_start,
+	.stop = da9063_wdt_stop,
+	.ping = da9063_wdt_ping,
+	.set_timeout = da9063_wdt_set_timeout,
+};
+
+static int da9063_wdt_probe(struct platform_device *pdev)
+{
+	int ret;
Quoted 8 lines

drivers/watchdog/da9063_wdt.c line 176

Quoted 5 lines
drivers/watchdog/da9063_wdt.c new file +5 -0
+	struct da9063 *da9063 = dev_get_drvdata(pdev->dev.parent);
+	struct da9063_watchdog *wdt;
+
+	if (!da9063)
+		return -EPROBE_DEFER;

Can this ever happen ? Isn't this driver instantiated from its parent, which would ensure that the parent's driver data is set ?

No this can't happen. I replaced it with BUG_ON and also introduced a BUG_ON check for dev.parent.

Quoted 11 lines
Quoted 11 lines
drivers/watchdog/da9063_wdt.c new file +10 -0
+
+	wdt = devm_kzalloc(&pdev->dev, sizeof(*wdt), GFP_KERNEL);
+	if (!wdt)
+		return -ENOMEM;
+
+	wdt->da9063 = da9063;
+	mutex_init(&wdt->lock);
+
+	wdt->wdtdev.info = &da9063_watchdog_info;
+	wdt->wdtdev.ops = &da9063_watchdog_ops;
Quoted 8 lines

drivers/watchdog/da9063_wdt.c line 192

Quoted 6 lines
drivers/watchdog/da9063_wdt.c new file +6 -0
+	wdt->wdtdev.min_timeout = DA9063_WDT_MIN_TIMEOUT;
+	wdt->wdtdev.max_timeout = DA9063_WDT_MAX_TIMEOUT;
+	wdt->wdtdev.timeout = DA9063_WDG_TIMEOUT;
+
+	if (IS_ENABLED(CONFIG_WATCHDOG_NOWAYOUT))
+		wdt->wdtdev.status = WATCHDOG_NOWAYOUT_INIT_STATUS;

You don't need the if() here. That is embedded in WATCHDOG_NOWAYOUT_INIT_STATUS.

Removed.

Thanks,

Markus

Signature · 11 lines
Pengutronix e.K.                           |                             |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20140814/0c823e14/attachment-0001.sig>
maintainer
#4 lore raw

PATCH v3 1/1 watchdog: Add DA906x PMIC watchdog driver.

Quoted 13 lines · Markus Pargmann
Quoted 4 lines
Quoted 4 lines
drivers/watchdog/da9063_wdt.c new file +4 -0
+
+static int da9063_wdt_probe(struct platform_device *pdev)
+{
+	int ret;
Quoted 8 lines

drivers/watchdog/da9063_wdt.c line 176

Quoted 5 lines
drivers/watchdog/da9063_wdt.c new file +5 -0
+	struct da9063 *da9063 = dev_get_drvdata(pdev->dev.parent);
+	struct da9063_watchdog *wdt;
+
+	if (!da9063)
+		return -EPROBE_DEFER;

Can this ever happen ? Isn't this driver instantiated from its parent, which would ensure that the parent's driver data is set ?

Quoted 3 lines

No this can't happen. I replaced it with BUG_ON and also introduced a BUG_ON check for dev.parent.

That is a bit too much. You should only use BUG_ON if the kernel can not continue to run, which should not be the case here. I would suggest to return -ENODEV.

Guenter

#5 lore raw

PATCH v3 1/1 watchdog: Add DA906x PMIC watchdog driver.

Quoted 15 lines · Guenter Roeck
Quoted 13 lines · Markus Pargmann
Quoted 12 lines

drivers/watchdog/da9063_wdt.c line 176

Quoted 9 lines
drivers/watchdog/da9063_wdt.c new file +9 -0
+
+static int da9063_wdt_probe(struct platform_device *pdev)
+{
+	int ret;
+	struct da9063 *da9063 = dev_get_drvdata(pdev->dev.parent);
+	struct da9063_watchdog *wdt;
+
+	if (!da9063)
+		return -EPROBE_DEFER;

Can this ever happen ? Isn't this driver instantiated from its parent, which would ensure that the parent's driver data is set ?

Quoted 6 lines
Quoted 3 lines

No this can't happen. I replaced it with BUG_ON and also introduced a BUG_ON check for dev.parent.

That is a bit too much. You should only use BUG_ON if the kernel can not continue to run, which should not be the case here. I would suggest to return -ENODEV.

Okay, thanks.

Markus

Signature · 11 lines
Pengutronix e.K.                           |                             |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20140815/9dfe2659/attachment.sig>