From: Hui Chun Ong <hui.chun.ong@ni.com>
To: "linux@roeck-us.net" <linux@roeck-us.net>,
"corbet@lwn.net" <corbet@lwn.net>,
"wim@iguana.be" <wim@iguana.be>
Cc: "rjw@rjwysocki.net" <rjw@rjwysocki.net>,
Julia Cartwright <julia.cartwright@ni.com>,
"mika.westerberg@linux.intel.com"
<mika.westerberg@linux.intel.com>,
"linux-watchdog@vger.kernel.org" <linux-watchdog@vger.kernel.org>,
"linux-doc@vger.kernel.org" <linux-doc@vger.kernel.org>,
Jonathan Hearn <jonathan.hearn@ni.com>
Subject: Re: [PATCH v2] watchdog: nic7018_wdt: Add NIC7018 watchdog driver
Date: Thu, 8 Dec 2016 09:19:22 +0000 [thread overview]
Message-ID: <1481188760.20527.16.camel@ni.com> (raw)
In-Reply-To: <ba6df233-af61-2553-27ab-72d5153c4f19@roeck-us.net>
On Thu, 2016-11-24 at 11:56 -0800, Guenter Roeck wrote:
> On 11/15/2016 07:21 PM, Hui Chun Ong wrote:
> >
> > Add support for the watchdog timer on PXI Embedded Controller.
> >
> > Signed-off-by: Hui Chun Ong <hui.chun.ong@ni.com>
> > ---
> > v1: Remove non-standard attributes.
> > Change from acpi_driver to platform_driver.
> > Rename driver from ni7018_wdt to nic7018_wdt.
> > ---
> > Documentation/watchdog/watchdog-parameters.txt | 5 +
> > drivers/watchdog/Kconfig | 10 +
> > drivers/watchdog/Makefile | 1 +
> > drivers/watchdog/nic7018_wdt.c | 303 +++++++++++++++++++++++++
> > 4 files changed, 319 insertions(+)
> > create mode 100644 drivers/watchdog/nic7018_wdt.c
> >
> > diff --git a/Documentation/watchdog/watchdog-parameters.txt b/Documentation/watchdog/watchdog-parameters.txt
> > index a8d3642..bd142fa 100644
> > --- a/Documentation/watchdog/watchdog-parameters.txt
> > +++ b/Documentation/watchdog/watchdog-parameters.txt
> > @@ -209,6 +209,11 @@ timeout: Initial watchdog timeout in seconds (0
> > nowayout: Watchdog cannot be stopped once started
> > (default=kernel config parameter)
> > -------------------------------------------------
> > +nic7018_wdt:
> > +timeout: Initial watchdog timeout in seconds (0
> > +nowayout: Watchdog cannot be stopped once started
> > + (default=kernel config parameter)
> > +-------------------------------------------------
> > nuc900_wdt:
> > heartbeat: Watchdog heartbeats in seconds.
> > (default = 15)
> > diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
> > index fdd3228..79fb224 100644
> > --- a/drivers/watchdog/Kconfig
> > +++ b/drivers/watchdog/Kconfig
> > @@ -1325,6 +1325,16 @@ config NI903X_WDT
> > To compile this driver as a module, choose M here: the module will be
> > called ni903x_wdt.
> >
> > +config NIC7018_WDT
> > + tristate "NIC7018 Watchdog"
> > + depends on X86 && ACPI
> > + select WATCHDOG_CORE
> > + ---help---
> > + This is the device driver for National Instruments NIC7018 Watchdog.
> > +
> This should describe what the watchdog is for.
>
> "Support for National Instruments NIC7018 Watchdog".
>
> >
> > + To compile this driver as a module, choose M here: the module will be
> > + called nic7018_wdt.
> > +
> > # M32R Architecture
> >
> > # M68K Architecture
> > diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
> > index caa9f4a..bd88e2e 100644
> > --- a/drivers/watchdog/Makefile
> > +++ b/drivers/watchdog/Makefile
> > @@ -139,6 +139,7 @@ obj-$(CONFIG_INTEL_SCU_WATCHDOG) += intel_scu_watchdog.o
> > obj-$(CONFIG_INTEL_MID_WATCHDOG) += intel-mid_wdt.o
> > obj-$(CONFIG_INTEL_MEI_WDT) += mei_wdt.o
> > obj-$(CONFIG_NI903X_WDT) += ni903x_wdt.o
> > +obj-$(CONFIG_NIC7018_WDT) += nic7018_wdt.o
> >
> > # M32R Architecture
> >
> > diff --git a/drivers/watchdog/nic7018_wdt.c b/drivers/watchdog/nic7018_wdt.c
> > new file mode 100644
> > index 0000000..780edc6
> > --- /dev/null
> > +++ b/drivers/watchdog/nic7018_wdt.c
> > @@ -0,0 +1,303 @@
> > +/*
> > + * Copyright (C) 2016 National Instruments Corp.
> > + *
> > + * 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.
> > + *
> > + * This program is distributed in the hope that it will be useful,
> > + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> > + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> > + * GNU General Public License for more details.
> > + */
> > +
> > +#include
> > +#include
> > +#include
> > +#include
> > +#include
> > +#include
> > +#include
> > +
> > +#define LOCK 0xA5
> > +#define UNLOCK 0x5A
> > +
> > +#define WDT_CTRL_RESET_EN BIT(7)
> > +#define WDT_RELOAD_PORT_EN BIT(7)
> > +
> Should include linux/bitops.h
>
> >
> > +#define WDT_CTRL 1
> > +#define WDT_RELOAD_CTRL 2
> > +#define WDT_PRESET_PRESCALE 4
> > +#define WDT_REG_LOCK 5
> > +#define WDT_COUNT 6
> > +#define WDT_RELOAD_PORT 7
> > +
> > +#define WDT_MIN_TIMEOUT 1
> > +#define WDT_MAX_TIMEOUT 464
> 32 * 15 would be 480 ? I assume the limit above is because of the rounding down
> by 16 in the driver, but that only shows that this rounding is maybe not the best
> idea. After all, the maximum timeout _is_ 480 seconds, if I understand the code
> correctly.
>
The formula for timeout calculation is = (period * counter) - (period / 2). So in this case, (32 * 15) - (32 / 2) = 464
> >
> > +#define WDT_DEFAULT_TIMEOUT 80
> > +
> Kind of an odd number, as it can not really be expressed by the hardware.
> The real timeout, if I understand the code correctly, would be 96 (32 * 3)
> seconds. Another data point that the rounding as implemented isn't necessarily
> the best possible implementation.
>
The timeout based on formula above is (32 * 3) - (32 / 16) = 80
> >
> > +#define WDT_IO_SIZE 8
> > +#define WDT_MAX_COUNTER 15
> > +
> > +static unsigned int timeout;
> > +module_param(timeout, uint, 0);
> > +MODULE_PARM_DESC(timeout,
> > + "Watchdog timeout in seconds. (default="
> > + __MODULE_STRING(WDT_DEFAULT_TIMEOUT) ")");
> > +
> > +static int nowayout = WATCHDOG_NOWAYOUT;
> > +module_param(nowayout, int, S_IRUGO);
> S_IRUGO is out of favor nowadays. Just use 0444.
> Sure you want this readable, though ? That is quite unusual.
>
> >
> > +MODULE_PARM_DESC(nowayout,
> > + "Watchdog cannot be stopped once started. (default="
> > + __MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
> > +
> > +struct nic7018_wdt {
> > + u16 io_base;
> > + u32 period_ms;
> > + struct mutex lock;
> This lock is not necessary. The watchdog core already provides locking.
>
> >
> > + struct watchdog_device wdd;
> > + struct platform_device *pdev;
> pdev isn't used anywhere.
>
> >
> > +};
> > +
> > +struct nic7018_config {
> > + u32 period_ms;
> > + u8 divider;
> > +};
> > +
> > +static const struct nic7018_config nic7018_configs[] = {
> > + { 125, 3 },
> > + { 2000, 4 },
> > + { 32000, 5 },
> > +};
> > +
> > +static inline u32 nic7018_timeout_ms(u32 period_ms, u8 counter)
> > +{
> > + return period_ms * counter - period_ms / 2;
> > +}
> > +
> > +static const struct nic7018_config *nic7018_get_config(u32 timeout_ms,
> > + u8 *counter)
> > +{
> > + u32 delta, i, best_delta = U32_MAX;
> > + const struct nic7018_config *config, *best_config = NULL;
> > + u8 count;
> > +
> > + for (i = 0; i < ARRAY_SIZE(nic7018_configs); i++) {
> > + config = &nic7018_configs[i];
> > +
> > + count = DIV_ROUND_UP(timeout_ms + config->period_ms / 2,
> > + config->period_ms);
> > +
> > + if (count > WDT_MAX_COUNTER)
> > + continue;
> > +
> > + delta = nic7018_timeout_ms(config->period_ms, count) -
> > + timeout_ms;
> > +
> > + if (delta < best_delta) {
> > + best_delta = delta;
> > + best_config = config;
> > + *counter = count;
> > + }
> > + }
> > + return best_config;
> > +}
> > +
> This appears overly complex. Ultimately it boils down to something like
>
> if (timeout_ms <= 125 * (WDT_MAX_COUNTER + 8) {
> *counter = min(WDT_MAX_COUNTER, DIV_ROUND_UP(timeout_ms, 125));
> return &config[0];
> } else if (timeout_ms <= 2000 * (WDT_MAX_COUNTER + 8) {
> *counter = min(WDT_MAX_COUNTER, DIV_ROUND_UP(timeout_ms, 2000));
> return &config[1];
> }
> *counter = min(WDT_MAX_COUNTER, DIV_ROUND_UP(timeout_ms, 32000));
> return &config[2];
>
> or, if you want to be fancy,
>
> for (i = 0; i < ARRAY_SIZE(nic7018_configs); i++) {
> config = &nic7018_configs[i];
> if (timeout_ms <= config->period_ms * (WDT_MAX_COUNTER + 8))
> break;
> }
> *counter = min(WDT_MAX_COUNTER, DIV_ROUND_UP(timeout_ms, config->period_ms));
> return config;
>
> >
> > +static int nic7018_set_timeout(struct watchdog_device *wdd,
> > + unsigned int timeout)
> > +{
> > + struct nic7018_wdt *wdt = watchdog_get_drvdata(wdd);
> > + const struct nic7018_config *config;
> > + u8 counter;
> > +
> > + config = nic7018_get_config(timeout * 1000, &counter);
> > + if (!config)
> > + return -EINVAL;
> This is not necessary. The maximum timeout guarantees that the core will not request
> a larger timeout than supported. Just have nic7018_get_config() never return an error
> (as suggested above).
>
> >
> > +
> > + mutex_lock(&wdt->lock);
> > + outb(counter << 4 | config->divider,
> > + wdt->io_base + WDT_PRESET_PRESCALE);
> > +
> > + wdd->timeout = nic7018_timeout_ms(config->period_ms, counter) / 1000;
> > + wdt->period_ms = config->period_ms;
> > + mutex_unlock(&wdt->lock);
> > +
> > + return 0;
> > +}
> > +
> > +static int nic7018_start(struct watchdog_device *wdd)
> > +{
> > + struct nic7018_wdt *wdt = watchdog_get_drvdata(wdd);
> > + u8 control;
> > +
> > + nic7018_set_timeout(wdd, wdd->timeout);
> > +
> Side note: As currently written, the above function can return an error
> (in theory). If that was the case, the code below would have more or less random
> results.
>
> >
> > + mutex_lock(&wdt->lock);
> > + control = inb(wdt->io_base + WDT_RELOAD_CTRL);
> > + outb(control | WDT_RELOAD_PORT_EN, wdt->io_base + WDT_RELOAD_CTRL);
> > +
> > + outb(1, wdt->io_base + WDT_RELOAD_PORT);
> > +
> > + control = inb(wdt->io_base + WDT_CTRL);
> > + outb(control | WDT_CTRL_RESET_EN, wdt->io_base + WDT_CTRL);
> > + mutex_unlock(&wdt->lock);
> > +
> > + return 0;
> > +}
> > +
> > +static int nic7018_stop(struct watchdog_device *wdd)
> > +{
> > + struct nic7018_wdt *wdt = watchdog_get_drvdata(wdd);
> > +
> > + mutex_lock(&wdt->lock);
> > + outb(0, wdt->io_base + WDT_CTRL);
> > + outb(0, wdt->io_base + WDT_RELOAD_CTRL);
> > + outb(0xF0, wdt->io_base + WDT_PRESET_PRESCALE);
> > + mutex_unlock(&wdt->lock);
> > +
> > + return 0;
> > +}
> > +
> > +static int nic7018_ping(struct watchdog_device *wdd)
> > +{
> > + struct nic7018_wdt *wdt = watchdog_get_drvdata(wdd);
> > +
> > + mutex_lock(&wdt->lock);
> > + outb(1, wdt->io_base + WDT_RELOAD_PORT);
> > + mutex_unlock(&wdt->lock);
> > +
> > + return 0;
> > +}
> > +
> > +static unsigned int nic7018_get_timeleft(struct watchdog_device *wdd)
> > +{
> > + struct nic7018_wdt *wdt = watchdog_get_drvdata(wdd);
> > + u8 count;
> > +
> > + mutex_lock(&wdt->lock);
> > + count = inb(wdt->io_base + WDT_COUNT) & 0xF;
> > + mutex_unlock(&wdt->lock);
> > +
> > + if (!count)
> > + return 0;
> > +
> This is only needed because of the rounding down in nic7018_timeout_ms().
> I am not really sure if that rounding adds any real value, especially
> since the result is then divided by 1000 without rounding. From my
> perspective, all it does is to add complexity. I would suggest to just return
> wdt->period_ms * count / 1000, or, if you want to be fancy,
> DIV_ROUND_CLOSEST(wdt->period_ms * count, 1000).
>
> >
> > + return nic7018_timeout_ms(wdt->period_ms, count) / 1000;
> > +}
> > +
> > +static const struct watchdog_info nic7018_wdd_info = {
> > + .options = WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING | WDIOF_MAGICCLOSE,
> > + .identity = "NIC7018 Watchdog",
> > +};
> > +
> > +static const struct watchdog_ops nic7018_wdd_ops = {
> > + .owner = THIS_MODULE,
> > + .start = nic7018_start,
> > + .stop = nic7018_stop,
> > + .ping = nic7018_ping,
> > + .set_timeout = nic7018_set_timeout,
> > + .get_timeleft = nic7018_get_timeleft,
> > +};
> > +
> > +static int nic7018_probe(struct platform_device *pdev)
> > +{
> > + struct device *dev = &pdev->dev;
> > + struct watchdog_device *wdd;
> > + struct nic7018_wdt *wdt;
> > + struct resource *io_rc;
> > + int ret;
> > +
> > + wdt = devm_kzalloc(dev, sizeof(*wdt), GFP_KERNEL);
> > + if (!wdt)
> > + return -ENOMEM;
> > +
> > + wdt->pdev = pdev;
> > + platform_set_drvdata(pdev, wdt);
> > +
> > + io_rc = platform_get_resource(pdev, IORESOURCE_IO, 0);
> > + if (!io_rc) {
> > + dev_err(dev, "missing IO resources\n");
> > + return -EINVAL;
> > + }
> > +
> > + if (resource_size(io_rc) < WDT_IO_SIZE) {
> > + dev_err(dev, "IO region too small\n");
> > + return -EINVAL;
> > + }
> > +
> It is highly unusual to check the resource size. I would suggest to drop it.
> Also, please use devm_ioremap_resource(), which returns an error code.
> With that, you don't have to check the error code from platform_get_resource();
> devm_ioremap_resource() does it for you.
> io_rc = platform_get_resource(pdev, IORESOURCE_IO, 0);
> wdt->io_base = devm_ioremap_resource(dev, res);
> if (IS_ERR(wdt->io_base))
> return PTR_ERR(wdt->io_base);
>
Not sure if I understand. Isn't "devm_ioremap_resource" only accept IORESOURCE_MEM not IORESOURCE_IO?
> >
> > + if (!devm_request_region(dev, io_rc->start, resource_size(io_rc),
> > + KBUILD_MODNAME)) {
> > + dev_err(dev, "failed to get IO region\n");
> > + return -EBUSY;
> > + }
> > +
> > + wdt->io_base = io_rc->start;
> > + mutex_init(&wdt->lock);
> > +
> > + wdd = &wdt->wdd;
> > + wdd->info = &nic7018_wdd_info;
> > + wdd->ops = &nic7018_wdd_ops;
> > + wdd->min_timeout = WDT_MIN_TIMEOUT;
> > + wdd->max_timeout = WDT_MAX_TIMEOUT;
> > + wdd->timeout = WDT_DEFAULT_TIMEOUT;
> > + wdd->parent = dev;
> > +
> > + watchdog_set_drvdata(wdd, wdt);
> > + watchdog_set_nowayout(wdd, nowayout);
> > +
> > + ret = watchdog_init_timeout(wdd, timeout, dev);
> > + if (ret)
> > + dev_err(dev, "unable to set timeout value, using default\n");
> If this is an error, the code whould abort. Either make it a warning, or abort.
>
> >
> > +
> > + /* Unlock WDT register */
> > + outb(UNLOCK, wdt->io_base + WDT_REG_LOCK);
> > +
> > + ret = watchdog_register_device(wdd);
> > + if (ret) {
> > + dev_err(dev, "failed to register watchdog\n");
> > + goto err_out;
> No output to WDT_REG_LOCK needed here ?
>
> >
> > + }
> > +
> > + dev_dbg(dev, "io_base=0x%04X, timeout=%d, nowayout=%d\n",
> > + wdt->io_base, timeout, nowayout);
> > + return 0;
> > +
> > +err_out:
> > + mutex_destroy(&wdt->lock);
> > + return ret;
> > +}
> > +
> > +static int nic7018_remove(struct platform_device *pdev)
> > +{
> > + struct nic7018_wdt *wdt = platform_get_drvdata(pdev);
> > +
> > + nic7018_stop(&wdt->wdd);
> > + watchdog_unregister_device(&wdt->wdd);
> > +
> > + /* Lock WDT register */
> > + outb(LOCK, wdt->io_base + WDT_REG_LOCK);
> > +
> > + mutex_destroy(&wdt->lock);
> > + return 0;
> > +}
> > +
> > +static const struct acpi_device_id nic7018_device_ids[] = {
> > + {"NIC7018", 0},
> > + {"", 0},
> > +};
> > +MODULE_DEVICE_TABLE(acpi, nic7018_device_ids);
> > +
> > +static struct platform_driver watchdog_driver = {
> > + .probe = nic7018_probe,
> > + .remove = nic7018_remove,
> > + .driver = {
> > + .name = KBUILD_MODNAME,
> > + .acpi_match_table = ACPI_PTR(nic7018_device_ids),
> > + },
> > +};
> > +
> > +module_platform_driver(watchdog_driver);
> > +
> > +MODULE_DESCRIPTION("National Instruments NIC7018 Watchdog driver");
> > +MODULE_AUTHOR("Hui Chun Ong <hui.chun.ong@ni.com>");
> > +MODULE_LICENSE("GPL");
> > --
> > 2.7.4
> >
> > --
> > 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
> >
next prev parent reply other threads:[~2016-12-08 9:19 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-11-16 3:21 [PATCH v2] watchdog: nic7018_wdt: Add NIC7018 watchdog driver Hui Chun Ong
2016-11-24 19:56 ` Guenter Roeck
2016-11-24 20:48 ` Guenter Roeck
2016-12-08 9:19 ` Hui Chun Ong [this message]
2016-12-08 14:15 ` Guenter Roeck
2016-12-09 4:07 ` Hui Chun Ong
2016-12-09 5:36 ` Guenter Roeck
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1481188760.20527.16.camel@ni.com \
--to=hui.chun.ong@ni.com \
--cc=corbet@lwn.net \
--cc=jonathan.hearn@ni.com \
--cc=julia.cartwright@ni.com \
--cc=linux-doc@vger.kernel.org \
--cc=linux-watchdog@vger.kernel.org \
--cc=linux@roeck-us.net \
--cc=mika.westerberg@linux.intel.com \
--cc=rjw@rjwysocki.net \
--cc=wim@iguana.be \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox