diff mbox series

[v4,04/11] watchdog: add support for sl28cpld watchdog

Message ID 20200604211039.12689-5-michael@walle.cc (mailing list archive)
State Not Applicable
Headers show
Series Add support for Kontron sl28cpld | expand

Commit Message

Michael Walle June 4, 2020, 9:10 p.m. UTC
Add support for the watchdog of the sl28cpld board management
controller. This is part of a multi-function device driver.

Signed-off-by: Michael Walle <michael@walle.cc>
Acked-by: Guenter Roeck <linux@roeck-us.net>
---
 drivers/watchdog/Kconfig        |  11 ++
 drivers/watchdog/Makefile       |   1 +
 drivers/watchdog/sl28cpld_wdt.c | 231 ++++++++++++++++++++++++++++++++
 3 files changed, 243 insertions(+)
 create mode 100644 drivers/watchdog/sl28cpld_wdt.c

Comments

Andy Shevchenko June 5, 2020, 8:14 a.m. UTC | #1
On Fri, Jun 5, 2020 at 12:14 AM Michael Walle <michael@walle.cc> wrote:
>
> Add support for the watchdog of the sl28cpld board management
> controller. This is part of a multi-function device driver.

...

> +#include <linux/of_device.h>

Didn't find a user of this.

...

> +static bool nowayout = WATCHDOG_NOWAYOUT;
> +module_param(nowayout, bool, 0);
> +MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default="
> +                               __MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
> +
> +static int timeout;
> +module_param(timeout, int, 0);
> +MODULE_PARM_DESC(timeout, "Initial watchdog timeout in seconds");

Guenter ACKed this, but I'm wondering why we still need module parameters...

...

> +       int ret;
> +
> +       ret = regmap_read(wdt->regmap, wdt->offset + WDT_COUNT, &val);
> +
> +       return (ret < 0) ? 0 : val;

Besides extra parentheses and questionable ' < 0' part, the following
would look better I think

ret = ...
if (ret)
  return 0;

return val;

...

> +       int ret;
> +
> +       ret = regmap_write(wdt->regmap, wdt->offset + WDT_TIMEOUT, timeout);
> +       if (!ret)
> +               wdd->timeout = timeout;
> +
> +       return ret;

Similar story here:

ret = ...
if (ret)
  return ret;

wdd->... = ...
return 0;

...

> +       ret = regmap_read(wdt->regmap, wdt->offset + WDT_CTRL, &status);

> +       if (ret < 0)

What ' < 0' means? Do we have some positive return values?
Ditto for all your code.

> +               return ret;

...

> +       if (status & WDT_CTRL_EN) {
> +               sl28cpld_wdt_start(wdd);

> +               set_bit(WDOG_HW_RUNNING, &wdd->status);

Do you need atomic op here? Why?

> +       }

...

> +static const struct of_device_id sl28cpld_wdt_of_match[] = {
> +       { .compatible = "kontron,sl28cpld-wdt" },

> +       {},

No comma.

> +};
Michael Walle June 5, 2020, 10:24 a.m. UTC | #2
Am 2020-06-05 10:14, schrieb Andy Shevchenko:
> On Fri, Jun 5, 2020 at 12:14 AM Michael Walle <michael@walle.cc> wrote:
>> 
>> Add support for the watchdog of the sl28cpld board management
>> controller. This is part of a multi-function device driver.
> 
> ...
> 
>> +#include <linux/of_device.h>
> 
> Didn't find a user of this.

I guess mod_devicetable.h then.

> 
> ...
> 
>> +static bool nowayout = WATCHDOG_NOWAYOUT;
>> +module_param(nowayout, bool, 0);
>> +MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started 
>> (default="
>> +                               __MODULE_STRING(WATCHDOG_NOWAYOUT) 
>> ")");
>> +
>> +static int timeout;
>> +module_param(timeout, int, 0);
>> +MODULE_PARM_DESC(timeout, "Initial watchdog timeout in seconds");
> 
> Guenter ACKed this, but I'm wondering why we still need module 
> parameters...

How would a user change the nowayout or the timeout? For the latter 
there is
a device tree entry, but thats not easy changable by the user.

> 
> ...
> 
>> +       int ret;
>> +
>> +       ret = regmap_read(wdt->regmap, wdt->offset + WDT_COUNT, &val);
>> +
>> +       return (ret < 0) ? 0 : val;
> 
> Besides extra parentheses and questionable ' < 0' part, the following
> would look better I think
> 
> ret = ...
> if (ret)
>   return 0;
> 
> return val;

yes, you're right.

> 
> ...
> 
>> +       int ret;
>> +
>> +       ret = regmap_write(wdt->regmap, wdt->offset + WDT_TIMEOUT, 
>> timeout);
>> +       if (!ret)
>> +               wdd->timeout = timeout;
>> +
>> +       return ret;
> 
> Similar story here:
> 
> ret = ...
> if (ret)
>   return ret;
> 
> wdd->... = ...
> return 0;
> 
> ...

ok

> 
>> +       ret = regmap_read(wdt->regmap, wdt->offset + WDT_CTRL, 
>> &status);
> 
>> +       if (ret < 0)
> 
> What ' < 0' means? Do we have some positive return values?
> Ditto for all your code.

probably not, I'll go over all return values and change them.

>> +               return ret;
> 
> ...
> 
>> +       if (status & WDT_CTRL_EN) {
>> +               sl28cpld_wdt_start(wdd);
> 
>> +               set_bit(WDOG_HW_RUNNING, &wdd->status);
> 
> Do you need atomic op here? Why?

I'd say consistency, all watchdog drivers do it like that. I just
had a look at where this is used, but it looks like access from
userspace is protected by a lock.

> 
>> +       }
> 
> ...
> 
>> +static const struct of_device_id sl28cpld_wdt_of_match[] = {
>> +       { .compatible = "kontron,sl28cpld-wdt" },
> 
>> +       {},
> 
> No comma.

yeah ;)
Andy Shevchenko June 5, 2020, 10:50 a.m. UTC | #3
On Fri, Jun 5, 2020 at 1:24 PM Michael Walle <michael@walle.cc> wrote:
> Am 2020-06-05 10:14, schrieb Andy Shevchenko:
> > On Fri, Jun 5, 2020 at 12:14 AM Michael Walle <michael@walle.cc> wrote:

...

> >> +static bool nowayout = WATCHDOG_NOWAYOUT;
> >> +module_param(nowayout, bool, 0);
> >> +MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started
> >> (default="
> >> +                               __MODULE_STRING(WATCHDOG_NOWAYOUT)
> >> ")");
> >> +
> >> +static int timeout;
> >> +module_param(timeout, int, 0);
> >> +MODULE_PARM_DESC(timeout, "Initial watchdog timeout in seconds");
> >
> > Guenter ACKed this, but I'm wondering why we still need module
> > parameters...
>
> How would a user change the nowayout or the timeout? For the latter
> there is
> a device tree entry, but thats not easy changable by the user.

Yes, it's more question to VIm and Guenter than to you.

...

> >> +       if (status & WDT_CTRL_EN) {
> >> +               sl28cpld_wdt_start(wdd);
> >
> >> +               set_bit(WDOG_HW_RUNNING, &wdd->status);
> >
> > Do you need atomic op here? Why?
>
> I'd say consistency, all watchdog drivers do it like that. I just
> had a look at where this is used, but it looks like access from
> userspace is protected by a lock.

Okay then.
Guenter Roeck June 5, 2020, 1:52 p.m. UTC | #4
On 6/5/20 3:50 AM, Andy Shevchenko wrote:
> On Fri, Jun 5, 2020 at 1:24 PM Michael Walle <michael@walle.cc> wrote:
>> Am 2020-06-05 10:14, schrieb Andy Shevchenko:
>>> On Fri, Jun 5, 2020 at 12:14 AM Michael Walle <michael@walle.cc> wrote:
> 
> ...
> 
>>>> +static bool nowayout = WATCHDOG_NOWAYOUT;
>>>> +module_param(nowayout, bool, 0);
>>>> +MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started
>>>> (default="
>>>> +                               __MODULE_STRING(WATCHDOG_NOWAYOUT)
>>>> ")");
>>>> +
>>>> +static int timeout;
>>>> +module_param(timeout, int, 0);
>>>> +MODULE_PARM_DESC(timeout, "Initial watchdog timeout in seconds");
>>>
>>> Guenter ACKed this, but I'm wondering why we still need module
>>> parameters...
>>
>> How would a user change the nowayout or the timeout? For the latter
>> there is
>> a device tree entry, but thats not easy changable by the user.
> 
> Yes, it's more question to VIm and Guenter than to you.
> 

Has support for providing module parameters with the kernel command line
been discontinued/deprecated, or did it run out of favor ? Sorry if I
missed that.

nowayout has a configuration default. A module parameter is sometimes
provided by drivers to be able to override it. The timeout provided
via devicetree or on the command line is only the initial/default
timeout, and the watchdog daemon can change it after opening the
watchdog device as it sees fit.

Guenter
Andy Shevchenko June 5, 2020, 2:09 p.m. UTC | #5
On Fri, Jun 05, 2020 at 06:52:00AM -0700, Guenter Roeck wrote:
> On 6/5/20 3:50 AM, Andy Shevchenko wrote:
> > On Fri, Jun 5, 2020 at 1:24 PM Michael Walle <michael@walle.cc> wrote:
> >> Am 2020-06-05 10:14, schrieb Andy Shevchenko:
> >>> On Fri, Jun 5, 2020 at 12:14 AM Michael Walle <michael@walle.cc> wrote:

...

> >>>> +static bool nowayout = WATCHDOG_NOWAYOUT;
> >>>> +module_param(nowayout, bool, 0);
> >>>> +MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started
> >>>> (default="
> >>>> +                               __MODULE_STRING(WATCHDOG_NOWAYOUT)
> >>>> ")");
> >>>> +
> >>>> +static int timeout;
> >>>> +module_param(timeout, int, 0);
> >>>> +MODULE_PARM_DESC(timeout, "Initial watchdog timeout in seconds");
> >>>
> >>> Guenter ACKed this, but I'm wondering why we still need module
> >>> parameters...
> >>
> >> How would a user change the nowayout or the timeout? For the latter
> >> there is
> >> a device tree entry, but thats not easy changable by the user.
> > 
> > Yes, it's more question to VIm and Guenter than to you.
> > 
> 
> Has support for providing module parameters with the kernel command line
> been discontinued/deprecated, or did it run out of favor ? Sorry if I
> missed that.

Latter according to Greg KH. One of the (plenty) examples [1].

[1]: https://www.mail-archive.com/driverdev-devel@linuxdriverproject.org/msg96495.html

> nowayout has a configuration default. A module parameter is sometimes
> provided by drivers to be able to override it. The timeout provided
> via devicetree or on the command line is only the initial/default
> timeout, and the watchdog daemon can change it after opening the
> watchdog device as it sees fit.

Thanks for explanation.
Guenter Roeck June 5, 2020, 3:05 p.m. UTC | #6
On Fri, Jun 05, 2020 at 05:09:11PM +0300, Andy Shevchenko wrote:
> On Fri, Jun 05, 2020 at 06:52:00AM -0700, Guenter Roeck wrote:
> > On 6/5/20 3:50 AM, Andy Shevchenko wrote:
> > > On Fri, Jun 5, 2020 at 1:24 PM Michael Walle <michael@walle.cc> wrote:
> > >> Am 2020-06-05 10:14, schrieb Andy Shevchenko:
> > >>> On Fri, Jun 5, 2020 at 12:14 AM Michael Walle <michael@walle.cc> wrote:
> 
> ...
> 
> > >>>> +static bool nowayout = WATCHDOG_NOWAYOUT;
> > >>>> +module_param(nowayout, bool, 0);
> > >>>> +MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started
> > >>>> (default="
> > >>>> +                               __MODULE_STRING(WATCHDOG_NOWAYOUT)
> > >>>> ")");
> > >>>> +
> > >>>> +static int timeout;
> > >>>> +module_param(timeout, int, 0);
> > >>>> +MODULE_PARM_DESC(timeout, "Initial watchdog timeout in seconds");
> > >>>
> > >>> Guenter ACKed this, but I'm wondering why we still need module
> > >>> parameters...
> > >>
> > >> How would a user change the nowayout or the timeout? For the latter
> > >> there is
> > >> a device tree entry, but thats not easy changable by the user.
> > > 
> > > Yes, it's more question to VIm and Guenter than to you.
> > > 
> > 
> > Has support for providing module parameters with the kernel command line
> > been discontinued/deprecated, or did it run out of favor ? Sorry if I
> > missed that.
> 
> Latter according to Greg KH. One of the (plenty) examples [1].
> 
> [1]: https://www.mail-archive.com/driverdev-devel@linuxdriverproject.org/msg96495.html
> 
What is the suggested replacement ?

Guenter
Andy Shevchenko June 5, 2020, 4:04 p.m. UTC | #7
On Fri, Jun 5, 2020 at 6:05 PM Guenter Roeck <linux@roeck-us.net> wrote:
> On Fri, Jun 05, 2020 at 05:09:11PM +0300, Andy Shevchenko wrote:
> > On Fri, Jun 05, 2020 at 06:52:00AM -0700, Guenter Roeck wrote:
> > > On 6/5/20 3:50 AM, Andy Shevchenko wrote:
> > > > On Fri, Jun 5, 2020 at 1:24 PM Michael Walle <michael@walle.cc> wrote:
> > > >> Am 2020-06-05 10:14, schrieb Andy Shevchenko:
> > > >>> On Fri, Jun 5, 2020 at 12:14 AM Michael Walle <michael@walle.cc> wrote:
> >
> > ...
> >
> > > >>>> +static bool nowayout = WATCHDOG_NOWAYOUT;
> > > >>>> +module_param(nowayout, bool, 0);
> > > >>>> +MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started
> > > >>>> (default="
> > > >>>> +                               __MODULE_STRING(WATCHDOG_NOWAYOUT)
> > > >>>> ")");
> > > >>>> +
> > > >>>> +static int timeout;
> > > >>>> +module_param(timeout, int, 0);
> > > >>>> +MODULE_PARM_DESC(timeout, "Initial watchdog timeout in seconds");
> > > >>>
> > > >>> Guenter ACKed this, but I'm wondering why we still need module
> > > >>> parameters...
> > > >>
> > > >> How would a user change the nowayout or the timeout? For the latter
> > > >> there is
> > > >> a device tree entry, but thats not easy changable by the user.
> > > >
> > > > Yes, it's more question to VIm and Guenter than to you.
> > > >
> > >
> > > Has support for providing module parameters with the kernel command line
> > > been discontinued/deprecated, or did it run out of favor ? Sorry if I
> > > missed that.
> >
> > Latter according to Greg KH. One of the (plenty) examples [1].
> >
> > [1]: https://www.mail-archive.com/driverdev-devel@linuxdriverproject.org/msg96495.html
> >
> What is the suggested replacement ?

In some mails he suggested to use sysfs.
Guenter Roeck June 5, 2020, 4:34 p.m. UTC | #8
On 6/5/20 9:04 AM, Andy Shevchenko wrote:
> On Fri, Jun 5, 2020 at 6:05 PM Guenter Roeck <linux@roeck-us.net> wrote:
>> On Fri, Jun 05, 2020 at 05:09:11PM +0300, Andy Shevchenko wrote:
>>> On Fri, Jun 05, 2020 at 06:52:00AM -0700, Guenter Roeck wrote:
>>>> On 6/5/20 3:50 AM, Andy Shevchenko wrote:
>>>>> On Fri, Jun 5, 2020 at 1:24 PM Michael Walle <michael@walle.cc> wrote:
>>>>>> Am 2020-06-05 10:14, schrieb Andy Shevchenko:
>>>>>>> On Fri, Jun 5, 2020 at 12:14 AM Michael Walle <michael@walle.cc> wrote:
>>>
>>> ...
>>>
>>>>>>>> +static bool nowayout = WATCHDOG_NOWAYOUT;
>>>>>>>> +module_param(nowayout, bool, 0);
>>>>>>>> +MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started
>>>>>>>> (default="
>>>>>>>> +                               __MODULE_STRING(WATCHDOG_NOWAYOUT)
>>>>>>>> ")");
>>>>>>>> +
>>>>>>>> +static int timeout;
>>>>>>>> +module_param(timeout, int, 0);
>>>>>>>> +MODULE_PARM_DESC(timeout, "Initial watchdog timeout in seconds");
>>>>>>>
>>>>>>> Guenter ACKed this, but I'm wondering why we still need module
>>>>>>> parameters...
>>>>>>
>>>>>> How would a user change the nowayout or the timeout? For the latter
>>>>>> there is
>>>>>> a device tree entry, but thats not easy changable by the user.
>>>>>
>>>>> Yes, it's more question to VIm and Guenter than to you.
>>>>>
>>>>
>>>> Has support for providing module parameters with the kernel command line
>>>> been discontinued/deprecated, or did it run out of favor ? Sorry if I
>>>> missed that.
>>>
>>> Latter according to Greg KH. One of the (plenty) examples [1].
>>>
>>> [1]: https://www.mail-archive.com/driverdev-devel@linuxdriverproject.org/msg96495.html
>>>
>> What is the suggested replacement ?
> 
> In some mails he suggested to use sysfs.
> 

Using sysfs specifically to defeat "nowayout" would defeat
its purpose, which is specifically to disable the possibility
to stop the watchdog. Using a module parameter (or rather boot
time parameter) is a bit different since that has to be done
consciously and is only valid for a given boot. The same is
true for the timeout parameter. If one can wait for sysfs (or
userspace) to be active, one can as well use the watchdog
daemon to update the timeout. The use case for the module
/ command line parameter or sysfs based timeout is to be
able to set it _before_ userspace is active.

So sysfs is not really an alternative.

Guenter
diff mbox series

Patch

diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
index 55b910c453da..2c7b0f10151e 100644
--- a/drivers/watchdog/Kconfig
+++ b/drivers/watchdog/Kconfig
@@ -340,6 +340,17 @@  config MLX_WDT
 	  To compile this driver as a module, choose M here: the
 	  module will be called mlx-wdt.
 
+config SL28CPLD_WATCHDOG
+	tristate "Kontron sl28 watchdog"
+	depends on MFD_SL28CPLD
+	select WATCHDOG_CORE
+	help
+	  Say Y here to include support for the watchdog timer
+	  on the Kontron sl28 CPLD.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called sl28cpld_wdt.
+
 # ALPHA Architecture
 
 # ARM Architecture
diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
index 97bed1d3d97c..aa6e41126901 100644
--- a/drivers/watchdog/Makefile
+++ b/drivers/watchdog/Makefile
@@ -225,3 +225,4 @@  obj-$(CONFIG_MENF21BMC_WATCHDOG) += menf21bmc_wdt.o
 obj-$(CONFIG_MENZ069_WATCHDOG) += menz69_wdt.o
 obj-$(CONFIG_RAVE_SP_WATCHDOG) += rave-sp-wdt.o
 obj-$(CONFIG_STPMIC1_WATCHDOG) += stpmic1_wdt.o
+obj-$(CONFIG_SL28CPLD_WATCHDOG) += sl28cpld_wdt.o
diff --git a/drivers/watchdog/sl28cpld_wdt.c b/drivers/watchdog/sl28cpld_wdt.c
new file mode 100644
index 000000000000..6c9518dc454a
--- /dev/null
+++ b/drivers/watchdog/sl28cpld_wdt.c
@@ -0,0 +1,231 @@ 
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * sl28cpld watchdog driver.
+ *
+ * Copyright 2019 Kontron Europe GmbH
+ */
+
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/of_device.h>
+#include <linux/platform_device.h>
+#include <linux/regmap.h>
+#include <linux/watchdog.h>
+
+/*
+ * Watchdog timer block registers.
+ */
+#define WDT_CTRL			0x00
+#define  WDT_CTRL_EN			BIT(0)
+#define  WDT_CTRL_LOCK			BIT(2)
+#define  WDT_CTRL_ASSERT_SYS_RESET	BIT(6)
+#define  WDT_CTRL_ASSERT_WDT_TIMEOUT	BIT(7)
+#define WDT_TIMEOUT			0x01
+#define WDT_KICK			0x02
+#define  WDT_KICK_VALUE			0x6b
+#define WDT_COUNT			0x03
+
+#define WDT_DEFAULT_TIMEOUT		10
+
+static bool nowayout = WATCHDOG_NOWAYOUT;
+module_param(nowayout, bool, 0);
+MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default="
+				__MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
+
+static int timeout;
+module_param(timeout, int, 0);
+MODULE_PARM_DESC(timeout, "Initial watchdog timeout in seconds");
+
+struct sl28cpld_wdt {
+	struct watchdog_device wdd;
+	struct regmap *regmap;
+	u32 offset;
+	bool assert_wdt_timeout;
+};
+
+static int sl28cpld_wdt_ping(struct watchdog_device *wdd)
+{
+	struct sl28cpld_wdt *wdt = watchdog_get_drvdata(wdd);
+
+	return regmap_write(wdt->regmap, wdt->offset + WDT_KICK,
+			    WDT_KICK_VALUE);
+}
+
+static int sl28cpld_wdt_start(struct watchdog_device *wdd)
+{
+	struct sl28cpld_wdt *wdt = watchdog_get_drvdata(wdd);
+	unsigned int val;
+
+	val = WDT_CTRL_EN | WDT_CTRL_ASSERT_SYS_RESET;
+	if (wdt->assert_wdt_timeout)
+		val |= WDT_CTRL_ASSERT_WDT_TIMEOUT;
+	if (nowayout)
+		val |= WDT_CTRL_LOCK;
+
+	return regmap_update_bits(wdt->regmap, wdt->offset + WDT_CTRL,
+				  val, val);
+}
+
+static int sl28cpld_wdt_stop(struct watchdog_device *wdd)
+{
+	struct sl28cpld_wdt *wdt = watchdog_get_drvdata(wdd);
+
+	return regmap_update_bits(wdt->regmap, wdt->offset + WDT_CTRL,
+				  WDT_CTRL_EN, 0);
+}
+
+static unsigned int sl28cpld_wdt_get_timeleft(struct watchdog_device *wdd)
+{
+	struct sl28cpld_wdt *wdt = watchdog_get_drvdata(wdd);
+	unsigned int val;
+	int ret;
+
+	ret = regmap_read(wdt->regmap, wdt->offset + WDT_COUNT, &val);
+
+	return (ret < 0) ? 0 : val;
+}
+
+static int sl28cpld_wdt_set_timeout(struct watchdog_device *wdd,
+				    unsigned int timeout)
+{
+	struct sl28cpld_wdt *wdt = watchdog_get_drvdata(wdd);
+	int ret;
+
+	ret = regmap_write(wdt->regmap, wdt->offset + WDT_TIMEOUT, timeout);
+	if (!ret)
+		wdd->timeout = timeout;
+
+	return ret;
+}
+
+static const struct watchdog_info sl28cpld_wdt_info = {
+	.options = WDIOF_MAGICCLOSE | WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING,
+	.identity = "sl28cpld watchdog",
+};
+
+static struct watchdog_ops sl28cpld_wdt_ops = {
+	.owner = THIS_MODULE,
+	.start = sl28cpld_wdt_start,
+	.stop = sl28cpld_wdt_stop,
+	.ping = sl28cpld_wdt_ping,
+	.set_timeout = sl28cpld_wdt_set_timeout,
+	.get_timeleft = sl28cpld_wdt_get_timeleft,
+};
+
+static int sl28cpld_wdt_probe(struct platform_device *pdev)
+{
+	struct watchdog_device *wdd;
+	struct sl28cpld_wdt *wdt;
+	unsigned int status;
+	unsigned int val;
+	int ret;
+
+	if (!pdev->dev.parent)
+		return -ENODEV;
+
+	wdt = devm_kzalloc(&pdev->dev, sizeof(*wdt), GFP_KERNEL);
+	if (!wdt)
+		return -ENOMEM;
+
+	wdt->regmap = dev_get_regmap(pdev->dev.parent, NULL);
+	if (!wdt->regmap)
+		return -ENODEV;
+
+	ret = device_property_read_u32(&pdev->dev, "reg", &wdt->offset);
+	if (ret)
+		return -EINVAL;
+
+	wdt->assert_wdt_timeout = device_property_read_bool(&pdev->dev,
+							    "kontron,assert-wdt-timeout-pin");
+
+	/* initialize struct watchdog_device */
+	wdd = &wdt->wdd;
+	wdd->parent = &pdev->dev;
+	wdd->info = &sl28cpld_wdt_info;
+	wdd->ops = &sl28cpld_wdt_ops;
+	wdd->min_timeout = 1;
+	wdd->max_timeout = 255;
+
+	watchdog_set_drvdata(wdd, wdt);
+	watchdog_stop_on_reboot(wdd);
+
+	/*
+	 * Read the status early, in case of an error, we haven't modified the
+	 * hardware.
+	 */
+	ret = regmap_read(wdt->regmap, wdt->offset + WDT_CTRL, &status);
+	if (ret < 0)
+		return ret;
+
+	/*
+	 * Initial timeout value, may be overwritten by device tree or module
+	 * parmeter in watchdog_init_timeout().
+	 *
+	 * Reading a zero here means that either the hardware has a default
+	 * value of zero (which is very unlikely and definitely a hardware
+	 * bug) or the bootloader set it to zero. In any case, we handle
+	 * this case gracefully and set out own timeout.
+	 */
+	ret = regmap_read(wdt->regmap, wdt->offset + WDT_TIMEOUT, &val);
+	if (ret < 0)
+		return ret;
+
+	if (val)
+		wdd->timeout = val;
+	else
+		wdd->timeout = WDT_DEFAULT_TIMEOUT;
+
+	watchdog_init_timeout(wdd, timeout, &pdev->dev);
+	sl28cpld_wdt_set_timeout(wdd, wdd->timeout);
+
+	/* if the watchdog is locked, we set nowayout */
+	if (status & WDT_CTRL_LOCK)
+		nowayout = true;
+	watchdog_set_nowayout(wdd, nowayout);
+
+	/*
+	 * If watchdog is already running, keep it enabled, but make
+	 * sure its mode is set correctly.
+	 */
+	if (status & WDT_CTRL_EN) {
+		sl28cpld_wdt_start(wdd);
+		set_bit(WDOG_HW_RUNNING, &wdd->status);
+	}
+
+	ret = devm_watchdog_register_device(&pdev->dev, wdd);
+	if (ret < 0) {
+		dev_err(&pdev->dev, "failed to register watchdog device\n");
+		return ret;
+	}
+
+	dev_info(&pdev->dev, "initial timeout %d sec%s\n",
+		 wdd->timeout, nowayout ? ", nowayout" : "");
+
+	return 0;
+}
+
+static const struct of_device_id sl28cpld_wdt_of_match[] = {
+	{ .compatible = "kontron,sl28cpld-wdt" },
+	{},
+};
+MODULE_DEVICE_TABLE(of, sl28cpld_wdt_of_match);
+
+static const struct platform_device_id sl28cpld_wdt_id_table[] = {
+	{ "sl28cpld-wdt" },
+	{},
+};
+MODULE_DEVICE_TABLE(platform, sl28cpld_wdt_id_table);
+
+static struct platform_driver sl28cpld_wdt_driver = {
+	.probe = sl28cpld_wdt_probe,
+	.id_table = sl28cpld_wdt_id_table,
+	.driver = {
+		.name = KBUILD_MODNAME,
+		.of_match_table = sl28cpld_wdt_of_match,
+	},
+};
+module_platform_driver(sl28cpld_wdt_driver);
+
+MODULE_DESCRIPTION("sl28cpld Watchdog Driver");
+MODULE_AUTHOR("Michael Walle <michael@walle.cc>");
+MODULE_LICENSE("GPL");