mbox series

[0/2] watchdog: Add Nuvoton NPCM (Poleg) driver

Message ID 20180302063746.10993-1-joel@jms.id.au
Headers show
Series watchdog: Add Nuvoton NPCM (Poleg) driver | expand

Message

Joel Stanley March 2, 2018, 6:37 a.m. UTC
This is a driver for the Poleg board that is in the process of being
upstreamed. I have tested it on an evaluation board.

The watchdog is a single register inside of the timer IP. This made me
think about how to describe the device tree bindings for a while, but
after some discussion I settled on describing the watchdog separately,
and giving it's own compatible string, and it's own driver.

The timer is being reviewed over here:

 https://lkml.org/lkml/2018/2/26/492

Joel Stanley (2):
  dt-bindings: watchdog: Add Nuvoton NPCM description
  watchdog: Add Nuvoton NPCM watchdog driver

 .../bindings/watchdog/nuvoton,npcm-wdt.txt         |  25 +++
 drivers/watchdog/Kconfig                           |  11 +
 drivers/watchdog/Makefile                          |   1 +
 drivers/watchdog/npcm_wdt.c                        | 223 +++++++++++++++++++++
 4 files changed, 260 insertions(+)
 create mode 100644 Documentation/devicetree/bindings/watchdog/nuvoton,npcm-wdt.txt
 create mode 100644 drivers/watchdog/npcm_wdt.c

-- 
2.15.1

--
To unsubscribe from this list: send the line "unsubscribe linux-watchdog" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Comments

Marcus Folkesson March 2, 2018, 11:20 a.m. UTC | #1
Hello Joel!

On Fri, Mar 02, 2018 at 05:07:46PM +1030, Joel Stanley wrote:
> The Nuvoton NPCM750 has a watchdog implemented as a single register

> inside the timer peripheral.

> 

> This driver exposes that watchdog as a standard watchdog device with

> coarse timeout intervals, limited by the combination of prescaler and

> counter that is provided by the hardware. The calculation is taken from

> the Nuvoton vendor tree.

> 

> There is a pre-timeout IRQ that is wired up. This timeout always occurs

> 1024 clocks before the timeout.

> 

> Signed-off-by: Joel Stanley <joel@jms.id.au>

> ---

>  drivers/watchdog/Kconfig    |  11 +++

>  drivers/watchdog/Makefile   |   1 +

>  drivers/watchdog/npcm_wdt.c | 223 ++++++++++++++++++++++++++++++++++++++++++++

>  3 files changed, 235 insertions(+)

>  create mode 100644 drivers/watchdog/npcm_wdt.c

> 

> diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig

> index aff773bcebdb..0c1cc68894e6 100644

> --- a/drivers/watchdog/Kconfig

> +++ b/drivers/watchdog/Kconfig

> @@ -513,6 +513,17 @@ config COH901327_WATCHDOG

>  	  This watchdog is used to reset the system and thus cannot be

>  	  compiled as a module.

>  

> +config NPCM7XX_WATCHDOG

> +	bool "NPCM750 watchdog"


Just asking, we do not want to make it possible to build this as a module?

> +	depends on ARCH_NPCM || COMPILE_TEST

> +	default y if ARCH_NPCM750

> +	select WATCHDOG_CORE

> +	help

> +	  Say Y here to include Watchdog timer support for the

> +	  watchdog embedded into the NPCM7xx.

> +	  This watchdog is used to reset the system and thus cannot be

> +	  compiled as a module.

> +

>  config TWL4030_WATCHDOG

>  	tristate "TWL4030 Watchdog"

>  	depends on TWL4030_CORE

> diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile

> index 0474d38aa854..97a5afb5cad2 100644

> --- a/drivers/watchdog/Makefile

> +++ b/drivers/watchdog/Makefile

> @@ -61,6 +61,7 @@ obj-$(CONFIG_ORION_WATCHDOG) += orion_wdt.o

>  obj-$(CONFIG_SUNXI_WATCHDOG) += sunxi_wdt.o

>  obj-$(CONFIG_RN5T618_WATCHDOG) += rn5t618_wdt.o

>  obj-$(CONFIG_COH901327_WATCHDOG) += coh901327_wdt.o

> +obj-$(CONFIG_NPCM7XX_WATCHDOG) += npcm_wdt.o

>  obj-$(CONFIG_STMP3XXX_RTC_WATCHDOG) += stmp3xxx_rtc_wdt.o

>  obj-$(CONFIG_NUC900_WATCHDOG) += nuc900_wdt.o

>  obj-$(CONFIG_TS4800_WATCHDOG) += ts4800_wdt.o

> diff --git a/drivers/watchdog/npcm_wdt.c b/drivers/watchdog/npcm_wdt.c

> new file mode 100644

> index 000000000000..71e3d4916514

> --- /dev/null

> +++ b/drivers/watchdog/npcm_wdt.c

> @@ -0,0 +1,223 @@

> +// SPDX-License-Identifier: GPL-2.0


The SPDX identifier `GPL-2.0` is GPLv2 *only*, your MODULE_LICENSE is
"GPLv2 or later". Please make the license info match.


> +// Copyright (c) 2018 Nuvoton Technology corporation.

> +// Copyright (c) 2018 IBM Corp.

> +

> +#include <linux/delay.h>

> +#include <linux/kernel.h>

> +#include <linux/module.h>

> +#include <linux/platform_device.h>

> +#include <linux/slab.h>

> +#include <linux/interrupt.h>

> +#include <linux/watchdog.h>

> +#include <asm/fiq.h>

> +#include <linux/of_irq.h>


#include <linux/bitopts.h>

Since you are using BIT()

See Documentation/process/submit-checklist.rst:
1) If you use a facility then #include the file that defines/declares
   that facility.  Don't depend on other header files pulling in ones
   that you use.


> +

> +#define NPCM_WTCR	0x1C

> +

> +#define NPCM_WTCLK	(BIT(10) | BIT(11))	/* Clock divider */

> +#define NPCM_WTE	BIT(7)			/* Enable */

> +#define NPCM_WTIE	BIT(6)			/* Enable irq */

> +#define NPCM_WTIS	(BIT(4) | BIT(5))	/* Interval selection */

> +#define NPCM_WTIF	BIT(3)			/* Interrupt flag*/

> +#define NPCM_WTRF	BIT(2)			/* Reset flag */

> +#define NPCM_WTRE	BIT(1)			/* Reset enable */

> +#define NPCM_WTR	BIT(0)			/* Reset counter */

> +

> +/*

> + * Watchdog timeouts

> + *

> + * 170     msec:    WTCLK=01 WTIS=00     VAL= 0x400

> + * 670     msec:    WTCLK=01 WTIS=01     VAL= 0x410

> + * 1360    msec:    WTCLK=10 WTIS=00     VAL= 0x800

> + * 2700    msec:    WTCLK=01 WTIS=10     VAL= 0x420

> + * 5360    msec:    WTCLK=10 WTIS=01     VAL= 0x810

> + * 10700   msec:    WTCLK=01 WTIS=11     VAL= 0x430

> + * 21600   msec:    WTCLK=10 WTIS=10     VAL= 0x820

> + * 43000   msec:    WTCLK=11 WTIS=00     VAL= 0xC00

> + * 85600   msec:    WTCLK=10 WTIS=11     VAL= 0x830

> + * 172000  msec:    WTCLK=11 WTIS=01     VAL= 0xC10

> + * 687000  msec:    WTCLK=11 WTIS=10     VAL= 0xC20

> + * 2750000 msec:    WTCLK=11 WTIS=11     VAL= 0xC30

> + */

> +

> +struct npcm_wdt {

> +	struct watchdog_device  wdd;

> +	struct device		*dev;

> +	void __iomem		*reg;

> +};

> +

> +static inline struct npcm_wdt *to_npcm_wdt(struct watchdog_device *wdd)

> +{

> +	return container_of(wdd, struct npcm_wdt, wdd);

> +}

> +

> +static int npcm_wdt_ping(struct watchdog_device *wdd)

> +{

> +	struct npcm_wdt *wdt = to_npcm_wdt(wdd);

> +	u32 val;

> +

> +	val = readl(wdt->reg);

> +	writel(val | NPCM_WTR, wdt->reg);

> +

> +	return 0;

> +}

> +

> +static int npcm_wdt_start(struct watchdog_device *wdd)

> +{

> +	struct npcm_wdt *wdt = to_npcm_wdt(wdd);

> +	u32 val;

> +

> +	val = NPCM_WTRE | NPCM_WTE | NPCM_WTR | NPCM_WTIE;

> +

> +	if (wdd->timeout < 2)

> +		val |= 0x800;

> +	else if (wdd->timeout < 3)

> +		val |= 0x420;

> +	else if (wdd->timeout < 6)

> +		val |= 0x810;

> +	else if (wdd->timeout < 11)

> +		val |= 0x430;

> +	else if (wdd->timeout < 22)

> +		val |= 0x820;

> +	else if (wdd->timeout < 44)

> +		val |= 0xC00;

> +	else if (wdd->timeout < 87)

> +		val |= 0x830;

> +	else if (wdd->timeout < 173)

> +		val |= 0xC10;

> +	else if (wdd->timeout < 688)

> +		val |= 0xC20;

> +	else if (wdd->timeout < 2751)

> +		val |= 0xC30;

> +	else

> +		val |= 0x830;

> +

> +	writel(val, wdt->reg);

> +

> +	return 0;

> +}

> +

> +static int npcm_wdt_stop(struct watchdog_device *wdd)

> +{

> +	struct npcm_wdt *wdt = to_npcm_wdt(wdd);

> +

> +	writel(0, wdt->reg);

> +

> +	return 0;

> +}

> +

> +

> +static int npcm_wdt_set_timeout(struct watchdog_device *wdd,

> +				unsigned int timeout)

> +{

> +	wdd->timeout = timeout;     /* New timeout */

> +	if (watchdog_active(wdd))

> +		npcm_wdt_start(wdd);

> +

> +	return 0;

> +}

> +

> +static irqreturn_t npcm_wdt_interrupt(int irq, void *data)

> +{

> +	struct npcm_wdt *wdt = data;

> +

> +	watchdog_notify_pretimeout(&wdt->wdd);

> +

> +	return IRQ_HANDLED;

> +}

> +

> +static int npcm_wdt_restart(struct watchdog_device *wdd,

> +			    unsigned long action, void *data)

> +{

> +	struct npcm_wdt *wdt = to_npcm_wdt(wdd);

> +

> +	writel(NPCM_WTR | NPCM_WTRE | NPCM_WTE, wdt->reg);

> +	udelay(1000);

> +

> +	return 0;

> +}

> +

> +static const struct watchdog_info npcm_wdt_info = {

> +	.identity	= KBUILD_MODNAME,

> +	.options	= WDIOF_SETTIMEOUT

> +			| WDIOF_KEEPALIVEPING

> +			| WDIOF_MAGICCLOSE,

> +};

> +

> +static const struct watchdog_ops npcm_wdt_ops = {

> +	.owner = THIS_MODULE,

> +	.start = npcm_wdt_start,

> +	.stop = npcm_wdt_stop,

> +	.ping = npcm_wdt_ping,

> +	.set_timeout = npcm_wdt_set_timeout,

> +	.restart = npcm_wdt_restart,

> +};

> +

> +static int npcm_wdt_probe(struct platform_device *pdev)

> +{

> +	struct device *dev = &pdev->dev;

> +	struct npcm_wdt *wdt;

> +	struct resource *res;

> +	int irq;

> +	int ret;

> +

> +	wdt = devm_kzalloc(&pdev->dev, sizeof(*wdt), GFP_KERNEL);

> +	if (!wdt)

> +		return -ENOMEM;

> +

> +	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);

> +	wdt->reg = devm_ioremap_resource(dev, res);

> +	if (IS_ERR(wdt->reg))

> +		return PTR_ERR(wdt->reg);

> +

> +	irq = platform_get_irq(pdev, 0);

> +	if (!irq)

> +		return -EINVAL;

> +

> +	wdt->dev = dev;

> +	wdt->wdd.info = &npcm_wdt_info;

> +	wdt->wdd.ops = &npcm_wdt_ops;

> +	wdt->wdd.min_timeout = 1;

> +	wdt->wdd.max_timeout = 2751;

> +	wdt->wdd.parent = dev;

> +

> +	wdt->wdd.timeout = 86;

> +	watchdog_init_timeout(&wdt->wdd, 0, dev);


watchdog_init_timeout() will also read the `timeout-sec` property from
the devicetree if present, which could be a good thing.

See drivers/watchdog/watchdog_core.c for implementation.

Just put it as an optional property in the dt-bindings.

> +

> +	ret = devm_request_irq(dev, irq, npcm_wdt_interrupt, 0,

> +			"watchdog", wdt);

> +	if (ret)

> +		return ret;

> +

> +	ret = devm_watchdog_register_device(dev, &wdt->wdd);

> +	if (ret) {

> +		dev_err(dev, "failed to register watchdog\n");

> +		return ret;

> +	}

> +

> +	platform_set_drvdata(pdev, wdt);


Do we need to set drvdata? I don't see that we are using it?

> +	dev_info(dev, "NPCM watchdog driver enabled\n");

> +

> +	return 0;

> +}

> +

> +#ifdef CONFIG_OF

> +static const struct of_device_id npcm_wdt_match[] = {

> +	{.compatible = "nuvoton,npcm750-wdt"},

> +	{},

> +};

> +MODULE_DEVICE_TABLE(of, npcm_wdt_match);

> +#endif

> +

> +static struct platform_driver npcm_wdt_driver = {

> +	.probe		= npcm_wdt_probe,

> +	.driver		= {

> +		.name	= "npcm-wdt",

> +		.of_match_table = of_match_ptr(npcm_wdt_match),

> +	},

> +};

> +module_platform_driver(npcm_wdt_driver);

> +

> +MODULE_AUTHOR("Joel Stanley");

> +MODULE_DESCRIPTION("Watchdog driver for NPCM");

> +MODULE_LICENSE("GPL");

					^^
Either "GPL v2" here or change SPDX identifier.

> -- 

> 2.15.1



Overall I think it looks good.

Thanks,
Marcus Folkesson
Joel Stanley March 5, 2018, 5:42 a.m. UTC | #2
On Fri, Mar 2, 2018 at 9:50 PM, Marcus Folkesson
<marcus.folkesson@gmail.com> wrote:
> Hello Joel!

>

> On Fri, Mar 02, 2018 at 05:07:46PM +1030, Joel Stanley wrote:

>> The Nuvoton NPCM750 has a watchdog implemented as a single register

>> inside the timer peripheral.

>>

>> This driver exposes that watchdog as a standard watchdog device with

>> coarse timeout intervals, limited by the combination of prescaler and

>> counter that is provided by the hardware. The calculation is taken from

>> the Nuvoton vendor tree.

>>

>> There is a pre-timeout IRQ that is wired up. This timeout always occurs

>> 1024 clocks before the timeout.

>>

>> Signed-off-by: Joel Stanley <joel@jms.id.au>

>> ---

>>  drivers/watchdog/Kconfig    |  11 +++

>>  drivers/watchdog/Makefile   |   1 +

>>  drivers/watchdog/npcm_wdt.c | 223 ++++++++++++++++++++++++++++++++++++++++++++

>>  3 files changed, 235 insertions(+)

>>  create mode 100644 drivers/watchdog/npcm_wdt.c

>>

>> diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig

>> index aff773bcebdb..0c1cc68894e6 100644

>> --- a/drivers/watchdog/Kconfig

>> +++ b/drivers/watchdog/Kconfig

>> @@ -513,6 +513,17 @@ config COH901327_WATCHDOG

>>         This watchdog is used to reset the system and thus cannot be

>>         compiled as a module.

>>

>> +config NPCM7XX_WATCHDOG

>> +     bool "NPCM750 watchdog"

>

> Just asking, we do not want to make it possible to build this as a module?


The watchdog provides the SoC with a way to reboot itself, so I would
expect all users of the kernel to have it built in in case the system
cannot find it's initrd or root file system, and therefore is unable
to load the watchdog module.

I am happy to change it if you would prefer.

Thanks for the review.

Cheers,

Joel
--
To unsubscribe from this list: send the line "unsubscribe linux-watchdog" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Joel Stanley March 5, 2018, 5:43 a.m. UTC | #3
On Sat, Mar 3, 2018 at 12:29 AM, Guenter Roeck <linux@roeck-us.net> wrote:
>> +static int npcm_wdt_start(struct watchdog_device *wdd)

>> +{

>> +       struct npcm_wdt *wdt = to_npcm_wdt(wdd);

>> +       u32 val;

>> +

>> +       val = NPCM_WTRE | NPCM_WTE | NPCM_WTR | NPCM_WTIE;

>> +

>> +       if (wdd->timeout < 2)

>> +               val |= 0x800;

>> +       else if (wdd->timeout < 3)

>> +               val |= 0x420;

>> +       else if (wdd->timeout < 6)

>> +               val |= 0x810;

>> +       else if (wdd->timeout < 11)

>> +               val |= 0x430;

>> +       else if (wdd->timeout < 22)

>> +               val |= 0x820;

>> +       else if (wdd->timeout < 44)

>> +               val |= 0xC00;

>> +       else if (wdd->timeout < 87)

>> +               val |= 0x830;

>> +       else if (wdd->timeout < 173)

>> +               val |= 0xC10;

>> +       else if (wdd->timeout < 688)

>> +               val |= 0xC20;

>> +       else if (wdd->timeout < 2751)

>> +               val |= 0xC30;

>> +       else

>> +               val |= 0x830;

>

>

> I am lost here. If the requested timeout is 2751, the actual timeout

> is set to 85.6 seconds. Why ?

>


I suspect this was the default, so the original code decided to fall
back to this value.

I think it would make more sense to use 2750.

Cheers,

Joel
--
To unsubscribe from this list: send the line "unsubscribe linux-watchdog" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html