[PATCH V2 05/10] MIPS: lantiq: add watchdog support

[PATCH V2 05/10] MIPS: lantiq: add watchdog support

[John Crispin]

This patch adds the driver for the watchdog found inside the Lantiq SoC family.

Changes in V2
* add comments to explain register access
* cleanup resource allocation
* cleanup clock handling
* whitespace fixes

Signed-off-by: John Crispin <blogic@openwrt.org>
Signed-off-by: Ralph Hempel <ralph.hempel@lantiq.com>
Cc: Wim Van Sebroeck <wim@iguana.be>
Cc: linux-mips@linux-mips.org
Cc: linux-watchdog@vger.kernel.org
---
 drivers/watchdog/Kconfig      |    6 +
 drivers/watchdog/Makefile     |    1 +
 drivers/watchdog/lantiq_wdt.c |  233 +++++++++++++++++++++++++++++++++++++++++
 3 files changed, 240 insertions(+), 0 deletions(-)
 create mode 100644 drivers/watchdog/lantiq_wdt.c

diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
index 2e2400e..c64f8c3 100644
--- a/drivers/watchdog/Kconfig
+++ b/drivers/watchdog/Kconfig
@@ -972,6 +972,12 @@ config BCM63XX_WDT
 	  To compile this driver as a loadable module, choose M here.
 	  The module will be called bcm63xx_wdt.
 
+config LANTIQ_WDT
+	tristate "Lantiq SoC watchdog"
+	depends on LANTIQ
+	help
+	  Hardware driver for the Lantiq SoC Watchdog Timer.
+
 # PARISC Architecture
 
 # POWERPC Architecture
diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
index dd77665..68299db 100644
--- a/drivers/watchdog/Makefile
+++ b/drivers/watchdog/Makefile
@@ -121,6 +121,7 @@ obj-$(CONFIG_AR7_WDT) += ar7_wdt.o
 obj-$(CONFIG_TXX9_WDT) += txx9wdt.o
 obj-$(CONFIG_OCTEON_WDT) += octeon-wdt.o
 octeon-wdt-y := octeon-wdt-main.o octeon-wdt-nmi.o
+obj-$(CONFIG_LANTIQ_WDT) += lantiq_wdt.o
 
 # PARISC Architecture
 
diff --git a/drivers/watchdog/lantiq_wdt.c b/drivers/watchdog/lantiq_wdt.c
new file mode 100644
index 0000000..8515c1f
--- /dev/null
+++ b/drivers/watchdog/lantiq_wdt.c
@@ -0,0 +1,233 @@
+/*
+ *  This program is free software; you can redistribute it and/or modify it
+ *  under the terms of the GNU General Public License version 2 as published
+ *  by the Free Software Foundation.
+ *
+ *  Copyright (C) 2010 John Crispin <blogic@openwrt.org>
+ *  Based on EP93xx wdt driver
+ */
+
+#include <linux/module.h>
+#include <linux/fs.h>
+#include <linux/miscdevice.h>
+#include <linux/watchdog.h>
+#include <linux/platform_device.h>
+#include <linux/uaccess.h>
+#include <linux/clk.h>
+#include <linux/io.h>
+
+#include <lantiq.h>
+
+/* Section 3.4 of the datasheet
+   The password sequence protects the WDT control register from unintended
+   write actions, which might cause malfunction of the WDT.
+
+   essentially the following two magic passwords need to be written to allow
+   io access to the wdt core */
+#define LTQ_WDT_PW1		0x00BE0000
+#define LTQ_WDT_PW2		0x00DC0000
+
+#define LTQ_WDT_CR		0x03F0	/* watchdog control register */
+#define LTQ_WDT_SR		0x03F8	/* watchdog status register */
+
+#define LTQ_WDT_SR_EN		(0x1 << 31) /* enable bit */
+#define LTQ_WDT_SR_PWD		(0x3 << 26) /* turn on power */
+#define LTQ_WDT_SR_CLKDIV	(0x3 << 24) /* turn on clock and set */
+					    /* divider to 0x40000 */
+#define LTQ_WDT_DIVIDER		0x40000
+#define LTQ_MAX_TIMEOUT		((1 << 16) - 1)	/* the reload field is 16 bit */
+
+#ifndef CONFIG_WATCHDOG_NOWAYOUT
+static int ltq_wdt_ok_to_close;
+#endif
+
+static int ltq_wdt_timeout = 30;
+static __iomem void *ltq_wdt_membase;
+static unsigned long ltq_io_region_clk_rate;
+
+static int
+ltq_wdt_enable(unsigned int timeout)
+{
+	timeout = ((timeout * (ltq_io_region_clk_rate / LTQ_WDT_DIVIDER))
+		+ 0x1000);
+	if (timeout > LTQ_MAX_TIMEOUT)
+		timeout = LTQ_MAX_TIMEOUT;
+
+	/* write the first paswword magic */
+	ltq_w32(LTQ_WDT_PW1, ltq_wdt_membase + LTQ_WDT_CR);
+	/* write the second magic plus the configuration and new timeout */
+	ltq_w32(LTQ_WDT_SR_EN | LTQ_WDT_SR_PWD | LTQ_WDT_SR_CLKDIV |
+		LTQ_WDT_PW2 | timeout, ltq_wdt_membase + LTQ_WDT_CR);
+	return 0;
+}
+
+static void
+ltq_wdt_disable(void)
+{
+#ifndef CONFIG_WATCHDOG_NOWAYOUT
+	ltq_wdt_ok_to_close = 0;
+#endif
+	/* write the first paswword magic */
+	ltq_w32(LTQ_WDT_PW1, ltq_wdt_membase + LTQ_WDT_CR);
+	/* write the second paswword magic with no config
+	   this turns the watchdog off */
+	ltq_w32(LTQ_WDT_PW2, ltq_wdt_membase + LTQ_WDT_CR);
+}
+
+static ssize_t
+ltq_wdt_write(struct file *file, const char __user *data,
+		size_t len, loff_t *ppos)
+{
+	size_t i;
+	if (!len)
+		return 0;
+#ifndef CONFIG_WATCHDOG_NOWAYOUT
+	for (i = 0; i != len; i++) {
+		char c;
+		if (get_user(c, data + i))
+			return -EFAULT;
+		if (c == 'V')
+			ltq_wdt_ok_to_close = 1;
+	}
+#endif
+	ltq_wdt_enable(ltq_wdt_timeout);
+	return len;
+}
+
+static struct watchdog_info ident = {
+	.options = WDIOF_MAGICCLOSE | WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING,
+	.identity = "ltq_wdt",
+};
+
+static long
+ltq_wdt_ioctl(struct file *file,
+		unsigned int cmd, unsigned long arg)
+{
+	int ret = -ENOTTY;
+	switch (cmd) {
+	case WDIOC_GETSUPPORT:
+		ret = copy_to_user((struct watchdog_info __user *)arg, &ident,
+				sizeof(ident)) ? -EFAULT : 0;
+		break;
+
+	case WDIOC_GETTIMEOUT:
+		ret = put_user(ltq_wdt_timeout, (int __user *)arg);
+		break;
+
+	case WDIOC_SETTIMEOUT:
+		ret = get_user(ltq_wdt_timeout, (int __user *)arg);
+		if (!ret)
+			ltq_wdt_enable(ltq_wdt_timeout);
+		break;
+
+	case WDIOC_KEEPALIVE:
+		ltq_wdt_enable(ltq_wdt_timeout);
+		ret = 0;
+		break;
+	}
+	return ret;
+}
+
+static int
+ltq_wdt_open(struct inode *inode, struct file *file)
+{
+	ltq_wdt_enable(ltq_wdt_timeout);
+	return nonseekable_open(inode, file);
+}
+
+static int
+ltq_wdt_release(struct inode *inode, struct file *file)
+{
+#ifndef CONFIG_WATCHDOG_NOWAYOUT
+	if (ltq_wdt_ok_to_close)
+		ltq_wdt_disable();
+	else
+#endif
+		printk(KERN_ERR "ltq_wdt: watchdog closed without warning\n");
+	return 0;
+}
+
+static const struct file_operations ltq_wdt_fops = {
+	.owner		= THIS_MODULE,
+	.write		= ltq_wdt_write,
+	.unlocked_ioctl	= ltq_wdt_ioctl,
+	.open		= ltq_wdt_open,
+	.release	= ltq_wdt_release,
+	.llseek		= no_llseek,
+};
+
+static struct miscdevice ltq_wdt_miscdev = {
+	.minor	= WATCHDOG_MINOR,
+	.name	= "watchdog",
+	.fops	= &ltq_wdt_fops,
+};
+
+static int
+ltq_wdt_probe(struct platform_device *pdev)
+{
+	struct resource *res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	struct clk *clk;
+	int ret = 0;
+	if (!res) {
+		dev_err(&pdev->dev, "cannot obtain I/O memory region");
+		return -ENOENT;
+	}
+	res = devm_request_mem_region(&pdev->dev, res->start,
+		resource_size(res), dev_name(&pdev->dev));
+	if (!res) {
+		dev_err(&pdev->dev, "cannot request I/O memory region");
+		return -EBUSY;
+	}
+	ltq_wdt_membase = devm_ioremap_nocache(&pdev->dev, res->start,
+		resource_size(res));
+	if (!ltq_wdt_membase) {
+		dev_err(&pdev->dev, "cannot remap I/O memory region\n");
+		return -ENOMEM;
+	}
+	/* we do not need to enable the clock as it is always running */
+	clk = clk_get(&pdev->dev, "io");
+	ltq_io_region_clk_rate = clk_get_rate(clk);
+	clk_put(clk);
+	ret = misc_register(&ltq_wdt_miscdev);
+	if (ret)
+		return 0;
+
+	return ret;
+}
+
+static int
+ltq_wdt_remove(struct platform_device *dev)
+{
+	ltq_wdt_disable();
+	misc_deregister(&ltq_wdt_miscdev);
+	return 0;
+}
+
+static struct platform_driver ltq_wdt_driver = {
+	.probe = ltq_wdt_probe,
+	.remove = ltq_wdt_remove,
+	.driver = {
+		.name = "ltq_wdt",
+		.owner = THIS_MODULE,
+	},
+};
+
+static int __init
+init_ltq_wdt(void)
+{
+	return platform_driver_register(&ltq_wdt_driver);
+}
+
+static void __exit
+exit_ltq_wdt(void)
+{
+	platform_driver_unregister(&ltq_wdt_driver);
+}
+
+module_init(init_ltq_wdt);
+module_exit(exit_ltq_wdt);
+
+MODULE_AUTHOR("John Crispin <blogic@openwrt.org>");
+MODULE_DESCRIPTION("Lantiq Watchdog");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_MISCDEV(WATCHDOG_MINOR);
-- 
1.7.2.3

Re: [PATCH V2 05/10] MIPS: lantiq: add watchdog support

[Sergei Shtylyov]

Hello.

On 01-03-2011 19:13, John Crispin wrote:

> This patch adds the driver for the watchdog found inside the Lantiq SoC family.

> Changes in V2
> * add comments to explain register access
> * cleanup resource allocation
> * cleanup clock handling
> * whitespace fixes

    The patch revision history is normally put under --- tearline.

> Signed-off-by: John Crispin<blogic@openwrt.org>
> Signed-off-by: Ralph Hempel<ralph.hempel@lantiq.com>
> Cc: Wim Van Sebroeck<wim@iguana.be>
> Cc: linux-mips@linux-mips.org
> Cc: linux-watchdog@vger.kernel.org
[...]

> diff --git a/drivers/watchdog/lantiq_wdt.c b/drivers/watchdog/lantiq_wdt.c
> new file mode 100644
> index 0000000..8515c1f
> --- /dev/null
> +++ b/drivers/watchdog/lantiq_wdt.c
> @@ -0,0 +1,233 @@
> +/*
> + *  This program is free software; you can redistribute it and/or modify it
> + *  under the terms of the GNU General Public License version 2 as published
> + *  by the Free Software Foundation.
> + *
> + *  Copyright (C) 2010 John Crispin<blogic@openwrt.org>
> + *  Based on EP93xx wdt driver
> + */
> +
> +#include<linux/module.h>
> +#include<linux/fs.h>
> +#include<linux/miscdevice.h>
> +#include<linux/watchdog.h>
> +#include<linux/platform_device.h>
> +#include<linux/uaccess.h>
> +#include<linux/clk.h>
> +#include<linux/io.h>
> +
> +#include<lantiq.h>
> +
> +/* Section 3.4 of the datasheet
> +   The password sequence protects the WDT control register from unintended
> +   write actions, which might cause malfunction of the WDT.
> +
> +   essentially the following two magic passwords need to be written to allow
> +   io access to the wdt core */

    The preferred style for the multi-line comments is this:

/*
  * bla
  * bla
  */

> +#define LTQ_WDT_CR		0x03F0	/* watchdog control register */
> +#define LTQ_WDT_SR		0x03F8	/* watchdog status register */
> +
> +#define LTQ_WDT_SR_EN		(0x1 << 31) /* enable bit */
> +#define LTQ_WDT_SR_PWD		(0x3 << 26) /* turn on power */
> +#define LTQ_WDT_SR_CLKDIV	(0x3 << 24) /* turn on clock and set */
> +					    /* divider to 0x40000 */
> +#define LTQ_WDT_DIVIDER		0x40000
> +#define LTQ_MAX_TIMEOUT		((1 << 16) - 1)	/* the reload field is 16 bit */

    Should align like the above...

> +
> +#ifndef CONFIG_WATCHDOG_NOWAYOUT
> +static int ltq_wdt_ok_to_close;
> +#endif
> +
> +static int ltq_wdt_timeout = 30;
> +static __iomem void *ltq_wdt_membase;

    It's normally "void __iomem *".

> +static unsigned long ltq_io_region_clk_rate;
> +
> +static int
> +ltq_wdt_enable(unsigned int timeout)
> +{
> +	timeout = ((timeout * (ltq_io_region_clk_rate / LTQ_WDT_DIVIDER))
> +		+ 0x1000);
> +	if (timeout > LTQ_MAX_TIMEOUT)
> +		timeout = LTQ_MAX_TIMEOUT;
> +
> +	/* write the first paswword magic */

    s/paswword/password/.

> +	ltq_w32(LTQ_WDT_PW1, ltq_wdt_membase + LTQ_WDT_CR);
> +	/* write the second magic plus the configuration and new timeout */
> +	ltq_w32(LTQ_WDT_SR_EN | LTQ_WDT_SR_PWD | LTQ_WDT_SR_CLKDIV |
> +		LTQ_WDT_PW2 | timeout, ltq_wdt_membase + LTQ_WDT_CR);
> +	return 0;

    This function should be *void* -- you alway return 0 and never check the 
result.

> +}
> +
> +static void
> +ltq_wdt_disable(void)
> +{
> +#ifndef CONFIG_WATCHDOG_NOWAYOUT
> +	ltq_wdt_ok_to_close = 0;
> +#endif
> +	/* write the first paswword magic */
> +	ltq_w32(LTQ_WDT_PW1, ltq_wdt_membase + LTQ_WDT_CR);
> +	/* write the second paswword magic with no config
> +	   this turns the watchdog off */

    Multi-line comment style...

> +	ltq_w32(LTQ_WDT_PW2, ltq_wdt_membase + LTQ_WDT_CR);
> +}
> +
> +static ssize_t
> +ltq_wdt_write(struct file *file, const char __user *data,
> +		size_t len, loff_t *ppos)
> +{
> +	size_t i;

    Empty line between the variables and code won't hurt...

> +	if (!len)
> +		return 0;
> +#ifndef CONFIG_WATCHDOG_NOWAYOUT
> +	for (i = 0; i != len; i++) {
> +		char c;

    Here too...

> +		if (get_user(c, data + i))
> +			return -EFAULT;
> +		if (c == 'V')
> +			ltq_wdt_ok_to_close = 1;
> +	}
> +#endif
[...]
> +static long
> +ltq_wdt_ioctl(struct file *file,
> +		unsigned int cmd, unsigned long arg)
> +{
> +	int ret = -ENOTTY;

    Empty line between the variables and code won't hurt...

> +	switch (cmd) {
> +	case WDIOC_GETSUPPORT:
> +		ret = copy_to_user((struct watchdog_info __user *)arg, &ident,
> +				sizeof(ident)) ? -EFAULT : 0;

    Doesn't copy_to_user() return 0 or -EFAULT?

> +static int
> +ltq_wdt_release(struct inode *inode, struct file *file)
> +{
> +#ifndef CONFIG_WATCHDOG_NOWAYOUT
> +	if (ltq_wdt_ok_to_close)
> +		ltq_wdt_disable();
> +	else
> +#endif
> +		printk(KERN_ERR "ltq_wdt: watchdog closed without warning\n");

    Use pr_err() instead.

> +static int
> +ltq_wdt_probe(struct platform_device *pdev)
> +{
> +	struct resource *res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> +	struct clk *clk;
> +	int ret = 0;

    Empty line between the variables and code won't hurt...

> +	/* we do not need to enable the clock as it is always running */
> +	clk = clk_get(&pdev->dev, "io");

    clk_get() may fail...

> +	ret = misc_register(&ltq_wdt_miscdev);
> +	if (ret)
> +		return 0;

    Er, didn't you mean:

	if (!ret)
		return 0;

> +	return ret;

    'ret' is always 0 here with your code.

> +}
> +
> +static int
> +ltq_wdt_remove(struct platform_device *dev)

    __exit?

> +static int __init
> +init_ltq_wdt(void)
> +{
> +	return platform_driver_register(&ltq_wdt_driver);

    Why not platfrom_driver_probe()? It's hardly a hotplug device...

WBR, Sergei

Re: [PATCH V2 05/10] MIPS: lantiq: add watchdog support

[Ralf Baechle]

On Wed, Mar 02, 2011 at 02:22:21PM +0300, Sergei Shtylyov wrote:

> >+	switch (cmd) {
> >+	case WDIOC_GETSUPPORT:
> >+		ret = copy_to_user((struct watchdog_info __user *)arg, &ident,
> >+				sizeof(ident)) ? -EFAULT : 0;
> 
>    Doesn't copy_to_user() return 0 or -EFAULT?

No and that's a common cause of bugs.  copy_{from,to}_user returns the
number of characters that could be be copied so the conversion to an
error code is needed here.

The function takes a void argument and there is no benefit from casting
to the full struct watchdog_info __user * pointer type other than maybe
clarity to the human reader.

While nitpicking - there should be one space between include and < in
#include <blah.h>.

  Ralf

Re: [PATCH V2 05/10] MIPS: lantiq: add watchdog support

[John Crispin]

Hi Ralf,
> While nitpicking - there should be one space between include and < in
> #include <blah.h>.
>
>   
where did you see that ?

thanks,
John

Re: [PATCH V2 05/10] MIPS: lantiq: add watchdog support

[Ralf Baechle]

On Wed, Mar 02, 2011 at 04:05:13PM +0100, John Crispin wrote:

> Hi Ralf,
> > While nitpicking - there should be one space between include and < in
> > #include <blah.h>.
> >
> >   
> where did you see that ?

> +#include<linux/module.h>
> +#include<linux/fs.h>
> +#include<linux/miscdevice.h>
> +#include<linux/watchdog.h>
> +#include<linux/platform_device.h>
> +#include<linux/uaccess.h>
> +#include<linux/clk.h>
> +#include<linux/io.h>
> +
> +#include<lantiq.h>

But that only seems to have happened to the code quoted in Sergei's mail.

  Ralf

Re: [PATCH V2 05/10] MIPS: lantiq: add watchdog support

[Sergei Shtylyov]

Hello.

Ralf Baechle wrote:

>> Hi Ralf,
>>> While nitpicking - there should be one space between include and < in
>>> #include <blah.h>.

>> where did you see that ?

>> +#include<linux/module.h>
>> +#include<linux/fs.h>
>> +#include<linux/miscdevice.h>
>> +#include<linux/watchdog.h>
>> +#include<linux/platform_device.h>
>> +#include<linux/uaccess.h>
>> +#include<linux/clk.h>
>> +#include<linux/io.h>
>> +
>> +#include<lantiq.h>

> But that only seems to have happened to the code quoted in Sergei's mail.

    Stupid Thunderbird is to be blamed here. :-)

>   Ralf

WBR, Sergei

Re: [PATCH V2 05/10] MIPS: lantiq: add watchdog support

[Sergei Shtylyov]

Hello.

Ralf Baechle wrote:

>>> +	switch (cmd) {
>>> +	case WDIOC_GETSUPPORT:
>>> +		ret = copy_to_user((struct watchdog_info __user *)arg, &ident,
>>> +				sizeof(ident)) ? -EFAULT : 0;

>>    Doesn't copy_to_user() return 0 or -EFAULT?

> No and that's a common cause of bugs.  copy_{from,to}_user returns the
> number of characters that could be be copied so the conversion to an
> error code is needed here.

    But then the code above would be wrong. Actually, it returns the number of 
bytes that could NOT be copied as I now see.

> The function takes a void argument and there is no benefit from casting
> to the full struct watchdog_info __user * pointer type other than maybe
> clarity to the human reader.

    Indeed.

>   Ralf

WBR, Sergei

Re: [PATCH V2 05/10] MIPS: lantiq: add watchdog support

[John Crispin]

Hi Sergei,

thanks for the comments, one question below
>> +    /* we do not need to enable the clock as it is always running */
>> +    clk = clk_get(&pdev->dev, "io");
>
>    clk_get() may fail...
>

lantiq socs have 2 static clock that are always running. so i think it
is safe to assume that this wont fail unless someone renames the clocks.

alternatively i could add a

if (!clk)
   BUG();

but i am not sure if it is required.

thanks,
John

Re: [PATCH V2 05/10] MIPS: lantiq: add watchdog support

[Jamie Iles]

On Wed, Mar 02, 2011 at 02:22:21PM +0300, Sergei Shtylyov wrote:
> >+static int
> >+ltq_wdt_remove(struct platform_device *dev)
> 
>    __exit?

No, I think this should be __devexit and the probe function should be 
__devinit.  When assigning the exit function to the platform_driver it 
should be surrounded with __devexit_p().

Jamie

Re: [PATCH V2 05/10] MIPS: lantiq: add watchdog support

[Sergei Shtylyov]

Hello.

On 03-03-2011 13:15, Jamie Iles wrote:

>>> +static int
>>> +ltq_wdt_remove(struct platform_device *dev)

>>     __exit?

> No, I think this should be __devexit and the probe function should be
> __devinit.  When assigning the exit function to the platform_driver it
> should be surrounded with __devexit_p().

    Why? Is WDT really a hotplug device?

> Jamie

WBR, Sergei