[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20130531114037.GA13533@roeck-us.net>
Date: Fri, 31 May 2013 04:40:37 -0700
From: Guenter Roeck <linux@...ck-us.net>
To: Johannes Thumshirn <johannes.thumshirn@....de>
Cc: wim@...ana.be, guenter@...ck-us.net,
linux-watchdog@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v6] watchdog: New watchdog driver for MEN A21 watchdogs
On Fri, May 31, 2013 at 01:08:52PM +0200, Johannes Thumshirn wrote:
> This patch adds the driver for the watchdog devices found on MEN Mikro
> Elektronik A21 VMEbus CPU Carrier Boards. It has DT-support and uses the
> watchdog framework.
>
> Revision 2:
> * Removed unneeded open flag in struct a21_wdt_drv
> * Corrected 3bit reason code from gpio
> * Additional sysfs files are now part of watchdog sysfs
> * Changed OFF/ON delay in ping from 400ms to 10ns
> * Reworked timeout setting
> * Removed a21_wdt_ioctl(...)
>
> Revision 3:
> * Changed pr_{err,info} to dev_{err,info}
> * Removed out of memory error print
> * Transition from "fast" to "slow" mode not allowed by chip
>
> Revision 4:
> * Remove reboot_notifier and place disable code into platform_device's shutdown function
> * Removed sysfs interface
>
> Revision 5:
>
> * Added setting of .bootstatus on driver init
> * Added initial timeout on driver init
>
> Revision 6:
> * Use watchdog_init_timeout() to initialize timeout
>
> Signed-off-by: Johannes Thumshirn <johannes.thumshirn@....de>
Couple of additional comments ... sorry, middle of night here and my thinking is
not at its best.
> ---
> MAINTAINERS | 6 ++
> drivers/watchdog/Kconfig | 8 ++
> drivers/watchdog/Makefile | 1 +
> drivers/watchdog/mena21_wdt.c | 234 +++++++++++++++++++++++++++++++++++++++++
> 4 files changed, 249 insertions(+)
> create mode 100644 drivers/watchdog/mena21_wdt.c
>
> diff --git a/MAINTAINERS b/MAINTAINERS
> index 7714c3c..023945a 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -5307,6 +5307,12 @@ F: drivers/mtd/
> F: include/linux/mtd/
> F: include/uapi/mtd/
>
> +MEN A21 WATCHDOG DRIVER
> +M: Johannes Thumshirn <johannes.thumshirn@....de>
> +L: linux-watchdog@...r.kernel.org
> +S: Supported
> +F: drivers/watchdog/mena21_wdt.c
> +
> METAG ARCHITECTURE
> M: James Hogan <james.hogan@...tec.com>
> S: Supported
> diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
> index e89fc31..192b84d 100644
> --- a/drivers/watchdog/Kconfig
> +++ b/drivers/watchdog/Kconfig
> @@ -1172,6 +1172,14 @@ config BOOKE_WDT_DEFAULT_TIMEOUT
>
> The value can be overridden by the wdt_period command-line parameter.
>
> +config MEN_A21_WDT
> + tristate "MEN A21 VME CPU Carrier Board Watchdog Timer"
> + select WATCHDOG_CORE
> + help
> + Watchdog driver for MEN A21 VMEbus CPU Carrier Boards.
> +
> + If unsure select N here.
> +
> # PPC64 Architecture
>
> config WATCHDOG_RTAS
> diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
> index a300b94..bffdcb1 100644
> --- a/drivers/watchdog/Makefile
> +++ b/drivers/watchdog/Makefile
> @@ -143,6 +143,7 @@ obj-$(CONFIG_8xxx_WDT) += mpc8xxx_wdt.o
> obj-$(CONFIG_MV64X60_WDT) += mv64x60_wdt.o
> obj-$(CONFIG_PIKA_WDT) += pika_wdt.o
> obj-$(CONFIG_BOOKE_WDT) += booke_wdt.o
> +obj-$(CONFIG_MEN_A21_WDT) += mena21_wdt.o
>
> # PPC64 Architecture
> obj-$(CONFIG_WATCHDOG_RTAS) += wdrtas.o
> diff --git a/drivers/watchdog/mena21_wdt.c b/drivers/watchdog/mena21_wdt.c
> new file mode 100644
> index 0000000..9874c63
> --- /dev/null
> +++ b/drivers/watchdog/mena21_wdt.c
> @@ -0,0 +1,234 @@
> +/*
> + * Watchdog driver for the A21 VME CPU Boards
> + *
> + * Copyright (C) 2013 MEN Mikro Elektronik Nuernberg GmbH
> + *
> + * 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
> + */
> +#include <linux/module.h>
> +#include <linux/moduleparam.h>
> +#include <linux/types.h>
> +#include <linux/kernel.h>
> +#include <linux/slab.h>
> +#include <linux/platform_device.h>
> +#include <linux/watchdog.h>
> +#include <linux/uaccess.h>
> +#include <linux/gpio.h>
> +#include <linux/delay.h>
> +#include <linux/bitops.h>
> +
> +#define GPIO_WD_ENAB 169
> +#define GPIO_WD_FAST 170
> +#define GPIO_WD_TRIG 171
> +
> +#define GPIO_RST_CAUSE_BASE 166
> +
I think I asked that before ... as you are supporting devicetree, gpio pins
should really be provided through devicetree properties and not be hardcoded.
> +struct a21_wdt_drv {
> + struct watchdog_device wdt;
> + struct mutex lock;
> +};
> +
> +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 a21_wdt_get_bootstatus(unsigned int *reset)
> +{
> + int i;
> +
> + for (i = 0; i < 3; i++)
> + *reset |= gpio_get_value(GPIO_RST_CAUSE_BASE + i)
> + ? (1 << i) : 0;
> +
> + if (*reset >= 8)
> + return -EINVAL;
> +
> + return 0;
> +}
> +
> +static int a21_wdt_start(struct watchdog_device *wdt)
> +{
> + struct a21_wdt_drv *drv = watchdog_get_drvdata(wdt);
> +
> + mutex_lock(&drv->lock);
> +
> + gpio_set_value(GPIO_WD_ENAB, 1);
> +
> + mutex_unlock(&drv->lock);
> +
> + return 0;
> +}
> +
> +static int a21_wdt_stop(struct watchdog_device *wdt)
> +{
> + struct a21_wdt_drv *drv = watchdog_get_drvdata(wdt);
> +
> + /* We don't stop if WDOG_NO_WAY_OUT is set */
> + if (test_bit(WDOG_NO_WAY_OUT, &wdt->status))
> + return -EINVAL;
> +
> + mutex_lock(&drv->lock);
> +
> + gpio_set_value(GPIO_WD_ENAB, 0);
> +
> + mutex_unlock(&drv->lock);
> +
> + return 0;
> +}
> +
> +static int a21_wdt_ping(struct watchdog_device *wdt)
> +{
> + struct a21_wdt_drv *drv = watchdog_get_drvdata(wdt);
> +
> + mutex_lock(&drv->lock);
> +
> + gpio_set_value(GPIO_WD_TRIG, 0);
> + ndelay(10);
> + gpio_set_value(GPIO_WD_TRIG, 1);
> +
> + mutex_unlock(&drv->lock);
> +
> + return 0;
> +}
> +
> +static int a21_wdt_set_timeout(struct watchdog_device *wdt,
> + unsigned int timeout)
> +{
> + struct a21_wdt_drv *drv = watchdog_get_drvdata(wdt);
> +
> + if (timeout != 1 && timeout != 30) {
> + dev_err(wdt->dev, "Only 1 and 30 allowed as timeout\n");
> + return -EINVAL;
> + }
> +
> + if (timeout == 30 && wdt->timeout == 1) {
> + dev_err(wdt->dev,
> + "Transition from fast to slow mode not allowed\n");
> + return -EINVAL;
> + }
> +
I dislike that [2 .. 29] timeout values are not supported, and would rather have
you accept that range and set the timeout to 30. Also, it is somewhat undesirable
that the timeout can not be changed back to 30 after being set to 1.
As Wim recommended, a softdog on top of the hardware watchdog would be the best
solution. I'll leave it up to him to decide if he wants to accept the code
as-is.
> + mutex_lock(&drv->lock);
> +
> + if (timeout == 1)
> + gpio_set_value(GPIO_WD_FAST, 1);
> + else
> + gpio_set_value(GPIO_WD_FAST, 0);
> +
> + wdt->timeout = timeout;
> +
> + mutex_unlock(&drv->lock);
> +
> + return 0;
> +}
> +
> +static const struct watchdog_info a21_wdt_info = {
> + .options = WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING | WDIOF_MAGICCLOSE,
> + .identity = "MEN A21 Watchdog",
> +};
> +
> +static const struct watchdog_ops a21_wdt_ops = {
> + .owner = THIS_MODULE,
> + .start = a21_wdt_start,
> + .stop = a21_wdt_stop,
> + .ping = a21_wdt_ping,
> + .set_timeout = a21_wdt_set_timeout,
> +};
> +
> +static struct watchdog_device a21_wdt = {
> + .info = &a21_wdt_info,
> + .ops = &a21_wdt_ops,
> + .min_timeout = 1,
> + .max_timeout = 30,
> +};
> +
> +static int a21_wdt_probe(struct platform_device *pdev)
> +{
> + struct a21_wdt_drv *drv;
> + unsigned int reset = 0;
> + int ret;
> +
> + dev_info(&pdev->dev, "MEN A21 watchdog timer driver enabled\n");
> +
> + drv = devm_kzalloc(&pdev->dev, sizeof(struct a21_wdt_drv), GFP_KERNEL);
> + if (!drv)
> + return -ENOMEM;
> +
> + mutex_init(&drv->lock);
> + watchdog_init_timeout(&a21_wdt, 30, &pdev->dev);
> + watchdog_set_nowayout(&a21_wdt, nowayout);
> + watchdog_set_drvdata(&a21_wdt, drv);
> +
> + ret = watchdog_register_device(&a21_wdt);
> + if (ret) {
> + dev_err(&pdev->dev, "Cannot register watchdog device\n");
> + goto err_register_wd;
> + }
> +
> + ret = a21_wdt_get_bootstatus(&reset);
> + if (ret)
> + dev_warn(&pdev->dev, "Reset Cause contains invalid data\n");
> + else {
> + if (reset == 2)
> + a21_wdt.bootstatus |= WDIOF_EXTERN1;
> + else if (reset == 4)
> + a21_wdt.bootstatus |= WDIOF_CARDRESET;
> + else if (reset == 5)
> + a21_wdt.bootstatus |= WDIOF_POWERUNDER;
> + else if (reset == 7)
> + a21_wdt.bootstatus |= WDIOF_EXTERN2;
What about other causes ? No useful match ?
I think bootstatus should be set prior to registering the watchdog device
to avoid race conditions where an application reads it prior to being set.
> + }
> +
> + dev_set_drvdata(&pdev->dev, drv);
> +
> + return 0;
> +
> +err_register_wd:
> + mutex_destroy(&drv->lock);
> +
> + return ret;
> +}
> +
> +static int a21_wdt_remove(struct platform_device *pdev)
> +{
> + struct a21_wdt_drv *drv = dev_get_drvdata(&pdev->dev);
> +
> + dev_warn(&pdev->dev,
> + "Unregistering A21 watchdog driver, board may reboot\n");
> +
> +
> + watchdog_unregister_device(&drv->wdt);
> +
> + mutex_destroy(&drv->lock);
> +
> + return 0;
> +}
> +
> +static void a21_wdt_shutdown(struct platform_device *pdev)
> +{
> + gpio_set_value(GPIO_WD_ENAB, 0);
> +}
> +
> +static const struct of_device_id a21_wdt_ids[] = {
> + { .compatible = "men,a021-wdt" },
> + { },
> +};
> +
> +static struct platform_driver a21_wdt_driver = {
> + .probe = a21_wdt_probe,
> + .remove = a21_wdt_remove,
> + .shutdown = a21_wdt_shutdown,
> + .driver = {
> + .name = "a21-watchdog",
> + .of_match_table = a21_wdt_ids,
> + },
> +};
> +
> +module_platform_driver(a21_wdt_driver);
> +
> +MODULE_AUTHOR("MEN Mikro Elektronik");
> +MODULE_DESCRIPTION("MEN A21 Watchdog");
> +MODULE_LICENSE("GPL");
> +MODULE_ALIAS("platform:a21-watchdog");
> --
> 1.7.9.5
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-watchdog" in
> the body of a message to majordomo@...r.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists