public inbox for linux-watchdog@vger.kernel.org
 help / color / mirror / Atom feed
From: Guenter Roeck <linux@roeck-us.net>
To: Oleksij Rempel <linux@rempel-privat.de>
Cc: linux-watchdog@vger.kernel.org, wim@iguana.be
Subject: Re: [PATCH] watchdog: add Alphascale asm9260-wdt driver
Date: Mon, 26 Oct 2015 19:31:32 -0700	[thread overview]
Message-ID: <20151027023132.GA1270@roeck-us.net> (raw)
In-Reply-To: <1442568212-15509-1-git-send-email-linux@rempel-privat.de>

On Fri, Sep 18, 2015 at 11:23:32AM +0200, Oleksij Rempel wrote:
> Add WD support for Alphascale asm9260 SoC. This driver
> provide support for different function modes:
> - HW mode to trigger SoC reset on timeout
> - SW mode do soft reset if needed
> - DEBUG mode
> 
> Optional support for stopping watchdog. If reset binding are not provided
> this driver will work in nowayout mode.

Why ?

> 
> Signed-off-by: Oleksij Rempel <linux@rempel-privat.de>
> ---
>  .../bindings/watchdog/alphascale-asm9260.txt       |  38 ++
>  drivers/watchdog/Kconfig                           |   9 +
>  drivers/watchdog/Makefile                          |   1 +
>  drivers/watchdog/asm9260_wdt.c                     | 405 +++++++++++++++++++++
>  4 files changed, 453 insertions(+)
>  create mode 100644 Documentation/devicetree/bindings/watchdog/alphascale-asm9260.txt
>  create mode 100644 drivers/watchdog/asm9260_wdt.c
> 
> diff --git a/Documentation/devicetree/bindings/watchdog/alphascale-asm9260.txt b/Documentation/devicetree/bindings/watchdog/alphascale-asm9260.txt
> new file mode 100644
> index 0000000..cbe388b
> --- /dev/null
> +++ b/Documentation/devicetree/bindings/watchdog/alphascale-asm9260.txt
> @@ -0,0 +1,38 @@
> +Alphascale asm9260 Watchdog timer
> +
> +Required properties:
> +
> +- compatible : should be "alphascale,asm9260-wdt".
> +- reg : Specifies base physical address and size of the registers.
> +- clocks : the clock feeding the watchdog timer.

There are two -> clocks.

> +	Needed if platform uses clocks. See clock-bindings.txt

The rest of the code suggests that the clocks are mandatory.
probe fails if asm9260_wdt_get_dt_clks() returns an error,
and the clock frequency would be 0.

> +- clock-names : should be set to
> +	"mod" - source for tick counter.
> +	"ahb" - ahb gate.
> +
> +Optional properties:
> +- resets : phandle pointing to the system reset controller with correct
> +	reset line index for watchdog controller reset. This propertie is
> +	required if you need to disable "nowayout" and it works only with
> +	CONFIG_WATCHDOG_NOWAYOUT=n.
> +- reset-names : should be set to "wdt_rst" if "resets" is used.
> +- timeout-sec : shall contain the default watchdog timeout in seconds,
> +	if unset, the default timeout is 30 seconds.
> +- alphascale,mode : tree modes are supported
> +	"hw" - hw reset (defaul).
> +	"sw" - sw reset.
> +	"debug" - no action is taken.
> +
> +Example:
> +
> +watchdog0: watchdog@80048000 {
> +	compatible = "alphascale,asm9260-wdt";
> +	reg = <0x80048000 0x10>;
> +	clocks = <&acc CLKID_SYS_WDT>, <&acc CLKID_AHB_WDT>;
> +	clock-names = "mod", "ahb";
> +	interrupts = <55>;
> +	resets = <&rst WDT_RESET>;
> +	reset-names = "wdt_rst";
> +	timeout-sec = <30>;
> +	alphascale,mode = "hw"

Seems to miss a ';'.

Did you get any feedback from dt maintainers on those bindings ?

> +};
> diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
> index c68edc1..cc5f675 100644
> --- a/drivers/watchdog/Kconfig
> +++ b/drivers/watchdog/Kconfig
> @@ -173,6 +173,15 @@ config ARM_SP805_WATCHDOG
>  	  ARM Primecell SP805 Watchdog timer. This will reboot your system when
>  	  the timeout is reached.
>  
> +config ASM9260_WATCHDOG
> +	tristate "Alphascale ASM9260 watchdog"
> +	depends on MACH_ASM9260
> +	depends on OF
> +	select WATCHDOG_CORE
> +	help
> +	  Watchdog timer embedded into Alphascale asm9260 chips. This will reboot your
> +	  system when the timeout is reached.
> +
>  config AT91RM9200_WATCHDOG
>  	tristate "AT91RM9200 watchdog"
>  	depends on SOC_AT91RM9200 && MFD_SYSCON
> diff --git a/drivers/watchdog/Makefile b/drivers/watchdog/Makefile
> index 0c616e3..bd7b0cd 100644
> --- a/drivers/watchdog/Makefile
> +++ b/drivers/watchdog/Makefile
> @@ -30,6 +30,7 @@ obj-$(CONFIG_USBPCWATCHDOG) += pcwd_usb.o
>  
>  # ARM Architecture
>  obj-$(CONFIG_ARM_SP805_WATCHDOG) += sp805_wdt.o
> +obj-$(CONFIG_ASM9260_WATCHDOG) += asm9260_wdt.o
>  obj-$(CONFIG_AT91RM9200_WATCHDOG) += at91rm9200_wdt.o
>  obj-$(CONFIG_AT91SAM9X_WATCHDOG) += at91sam9_wdt.o
>  obj-$(CONFIG_CADENCE_WATCHDOG) += cadence_wdt.o
> diff --git a/drivers/watchdog/asm9260_wdt.c b/drivers/watchdog/asm9260_wdt.c
> new file mode 100644
> index 0000000..e5f859b
> --- /dev/null
> +++ b/drivers/watchdog/asm9260_wdt.c
> @@ -0,0 +1,405 @@
> +/*
> + * Watchdog driver for Alphascale ASM9260.
> + *
> + * Copyright (c) 2014 Oleksij Rempel <linux@rempel-privat.de>
> + *
> + * Licensed under GPLv2 or later.
> + */
> +
> +#include <linux/delay.h>
> +#include <linux/clk.h>
> +#include <linux/interrupt.h>
> +#include <linux/io.h>
> +#include <linux/module.h>
> +#include <linux/moduleparam.h>
> +#include <linux/of.h>
> +#include <linux/platform_device.h>
> +#include <linux/reboot.h>
> +#include <linux/reset.h>
> +#include <linux/uaccess.h>
> +#include <linux/watchdog.h>
> +
> +#define CLOCK_FREQ	1000000
> +
> +/* Watchdog Mode register */
> +#define HW_WDMOD			0x00
> +/* Wake interrupt. Set by HW, can't be cleared. */
> +#define BM_MOD_WDINT			BIT(3)

When using BIT macros, please include bitops.h.

> +/* This bit set if timeout reached. Cleared by SW. */
> +#define BM_MOD_WDTOF			BIT(2)
> +/* HW Reset on timeout */
> +#define BM_MOD_WDRESET			BIT(1)
> +/* WD enable */
> +#define BM_MOD_WDEN			BIT(0)
> +
> +/*
> + * Watchdog Timer Constant register
> + * Minimal value is 0xff, the meaning of this value
> + * depends on used clock: T = WDCLK * (0xff + 1) * 4
> + */
> +#define HW_WDTC				0x04
> +#define BM_WDTC_MAX(freq)		(0x7fffffff / (freq))
> +
> +/* Watchdog Feed register */
> +#define HW_WDFEED			0x08
> +
> +/* Watchdog Timer Value register */
> +#define HW_WDTV				0x0c
> +
> +#define ASM9260_WDT_DEFAULT_TIMEOUT	30
> +
> +enum asm9260_wdt_mode {
> +	HW_RESET,
> +	SW_RESET,
> +	DEBUG,
> +};
> +
> +struct asm9260_wdt_priv {
> +	struct device		*dev;
> +	struct watchdog_device	wdd;
> +	struct clk		*clk;
> +	struct clk		*clk_ahb;
> +	struct reset_control	*rst;
> +	struct notifier_block	restart_handler;
> +
> +	void __iomem		*iobase;
> +	int			irq;
> +	unsigned long		wdt_freq;
> +	enum asm9260_wdt_mode	mode;
> +};
> +
> +static int asm9260_wdt_feed(struct watchdog_device *wdd)
> +{
> +	struct asm9260_wdt_priv *priv = watchdog_get_drvdata(wdd);
> +
> +	iowrite32(0xaa, priv->iobase + HW_WDFEED);
> +	iowrite32(0x55, priv->iobase + HW_WDFEED);
> +
> +	return 0;
> +}
> +
> +static unsigned int asm9260_wdt_gettimeleft(struct watchdog_device *wdd)
> +{
> +	struct asm9260_wdt_priv *priv = watchdog_get_drvdata(wdd);
> +	u32 counter;
> +
> +	counter = ioread32(priv->iobase + HW_WDTV);
> +
> +	return DIV_ROUND_CLOSEST(counter, priv->wdt_freq);
> +}
> +
> +static int asm9260_wdt_updatetimeout(struct watchdog_device *wdd)
> +{
> +	struct asm9260_wdt_priv *priv = watchdog_get_drvdata(wdd);
> +	u32 counter;
> +
> +	counter = wdd->timeout * priv->wdt_freq;
> +
> +	iowrite32(counter, priv->iobase + HW_WDTC);
> +
> +	return 0;
> +}
> +
> +static int asm9260_wdt_enable(struct watchdog_device *wdd)
> +{
> +	struct asm9260_wdt_priv *priv = watchdog_get_drvdata(wdd);
> +	u32 mode = 0;
> +
> +	if (priv->mode == HW_RESET)
> +		mode = BM_MOD_WDRESET;
> +
> +	iowrite32(BM_MOD_WDEN | mode, priv->iobase + HW_WDMOD);
> +
> +	asm9260_wdt_updatetimeout(wdd);
> +
> +	asm9260_wdt_feed(wdd);
> +
> +	return 0;
> +}
> +
> +static int asm9260_wdt_disable(struct watchdog_device *wdd)
> +{
> +	struct asm9260_wdt_priv *priv = watchdog_get_drvdata(wdd);
> +
> +	/* The only way to disable WD is to reset it. */
> +	reset_control_assert(priv->rst);
> +	reset_control_deassert(priv->rst);
> +
> +	return 0;
> +}
> +
> +static int asm9260_wdt_settimeout(struct watchdog_device *wdd, unsigned int to)
> +{
> +	wdd->timeout = to;
> +	asm9260_wdt_updatetimeout(wdd);
> +
> +	return 0;
> +}
> +
> +static void asm9260_wdt_sys_reset(struct asm9260_wdt_priv *priv)
> +{
> +	/* init WD if it was not started */
> +	priv->wdd.timeout = 1;
> +
> +	iowrite32(BM_MOD_WDEN | BM_MOD_WDRESET, priv->iobase + HW_WDMOD);
> +
> +	asm9260_wdt_updatetimeout(&priv->wdd);
> +	/* first pass correct sequence */
> +	asm9260_wdt_feed(&priv->wdd);
> +	/*
> +	 * Then write wrong pattern to the feed to trigger reset
> +	 * ASAP.
> +	 */
> +	iowrite32(0xff, priv->iobase + HW_WDFEED);
> +
> +	mdelay(1000);
> +}
> +
> +static irqreturn_t asm9260_wdt_irq(int irq, void *devid)
> +{
> +	struct asm9260_wdt_priv *priv = devid;
> +	u32 stat;
> +
> +	stat = ioread32(priv->iobase + HW_WDMOD);
> +	if (!(stat & BM_MOD_WDINT))
> +		return IRQ_NONE;
> +
> +	/*
> +	 * BM_MOD_WDINT flag can't be removed by SW. Only way is
> +	 * to reset WD Controller.
> +	 * TODO: add Controller reset if needed.

Is it needed or not ?

> +	 */
> +	if (priv->mode == DEBUG)
> +		dev_info(priv->dev, "Watchdog Timeout. Do nothing.\n");
> +	else {
> +		dev_info(priv->dev, "Watchdog Timeout. Doing SW Reset.\n");
> +		asm9260_wdt_sys_reset(priv);
> +	}
> +
> +	return IRQ_HANDLED;
> +}
> +
> +static int asm9260_restart_handler(struct notifier_block *this,
> +				unsigned long mode, void *cmd)
> +{
> +	struct asm9260_wdt_priv *priv =
> +		container_of(this, struct asm9260_wdt_priv, restart_handler);
> +
> +	asm9260_wdt_sys_reset(priv);
> +
> +	return NOTIFY_DONE;
> +}
> +
> +static const struct watchdog_info asm9260_wdt_ident = {
> +	.options          =     WDIOF_SETTIMEOUT | WDIOF_KEEPALIVEPING
> +				| WDIOF_MAGICCLOSE,
> +	.firmware_version =	0,

Assigning 0 to a static variable is not necessary.

> +	.identity         =	"Alphascale asm9260 Watchdog",
> +};
> +
> +static struct watchdog_ops asm9260_wdt_ops = {
> +	.owner		= THIS_MODULE,
> +	.start		= asm9260_wdt_enable,
> +	.stop		= asm9260_wdt_disable,
> +	.get_timeleft	= asm9260_wdt_gettimeleft,
> +	.ping		= asm9260_wdt_feed,
> +	.set_timeout	= asm9260_wdt_settimeout,
> +};
> +
> +static int __init asm9260_wdt_get_dt_clks(struct asm9260_wdt_priv *priv)
> +{
> +	int clk_idx = 0, err;
> +
> +	priv->clk = devm_clk_get(priv->dev, "mod");
> +	if (IS_ERR(priv->clk))
> +		goto out_err;
> +
> +	/* configure AHB clock */
> +	clk_idx = 1;
> +	priv->clk_ahb = devm_clk_get(priv->dev, "ahb");
> +	if (IS_ERR(priv->clk_ahb))
> +		goto out_err;
> +
> +	err = clk_prepare_enable(priv->clk_ahb);
> +	if (err)
> +		dev_err(priv->dev, "Failed to enable ahb_clk!\n");

Not really clear why you ignore those errors. Please explain.

> +
> +	err = clk_set_rate(priv->clk, CLOCK_FREQ);
> +	if (err) {
> +		clk_disable_unprepare(priv->clk_ahb);
> +		dev_err(priv->dev, "Failed to set rate!\n");
> +	}
> +
> +	err = clk_prepare_enable(priv->clk);
> +	if (err) {
> +		clk_disable_unprepare(priv->clk_ahb);
> +		dev_err(priv->dev, "Failed to enable clk!\n");
> +	}
> +
> +	/* wdt has internal divider */
> +	priv->wdt_freq = clk_get_rate(priv->clk) / 2;

Can wdt_freq ever be 0 ?

... yes, if any of the prepare_enable functions failed.


> +
> +	return 0;
> +out_err:
> +	dev_err(priv->dev, "%s: Failed to get clk (%i)\n", __func__, clk_idx);

"clk_idx" does not really provide useful information to the user.
It might make sense to use a better description where the error
occurs, and drop clk_idx.

> +	return 1;

This is not a valid error code.

> +}
> +
> +static void __init asm9260_wdt_get_dt_mode(struct asm9260_wdt_priv *priv)
> +{
> +	const char *tmp;
> +	int ret;
> +
> +	/* default mode */
> +	priv->mode = HW_RESET;
> +
> +	ret = of_property_read_string(priv->dev->of_node,
> +			"alphascale,mode", &tmp);
> +	if (ret)
> +		return;
> +
> +	if (!strcmp(tmp, "hw"))
> +		priv->mode = HW_RESET;
> +	else if (!strcmp(tmp, "sw"))
> +		priv->mode = SW_RESET;
> +	else if (!strcmp(tmp, "debug"))
> +		priv->mode = DEBUG;
> +	else {
> +		dev_warn(priv->dev, "unknown reset-type: %s. Using defaul \"hw\" mode.",
> +			tmp);

Returning -EINVAL would be more appropriate here.

> +		return;
> +	}
> +
> +	dev_info(priv->dev, "using \"%s\" mode.", tmp);

Is this noise really necessary ?

If anything, a single message at the end of the probe function
would be easier to accept.

> +}
> +
> +static int __init asm9260_wdt_probe(struct platform_device *pdev)
> +{
> +	struct asm9260_wdt_priv *priv;
> +	struct watchdog_device *wdd;
> +	struct resource *res;
> +	bool nowayout = WATCHDOG_NOWAYOUT;
> +	int ret;
> +
> +	priv = devm_kzalloc(&pdev->dev, sizeof(struct asm9260_wdt_priv),
> +			GFP_KERNEL);
> +	if (!priv)
> +		return -ENOMEM;
> +
> +	priv->dev = &pdev->dev;
> +
> +	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> +	priv->iobase = devm_ioremap_resource(&pdev->dev, res);
> +	if (IS_ERR(priv->iobase))
> +		return PTR_ERR(priv->iobase);
> +
> +	ret = asm9260_wdt_get_dt_clks(priv);
> +	if (ret)
> +		return ret;
> +
> +	priv->rst = devm_reset_control_get(&pdev->dev, "wdt_rst");
> +	if (IS_ERR(priv->rst)) {
> +		dev_info(&pdev->dev, "No reset control found. Enabling \"nowayout\" mode\n");

Noise. This should be obvious from the devicetree description.
I don't see value in repeating devicetree data.

> +		nowayout = 1;

Again, why ? nowayout should be independent.

> +		priv->rst = NULL;
> +	}
> +
> +	wdd = &priv->wdd;
> +	wdd->info = &asm9260_wdt_ident;
> +	wdd->ops = &asm9260_wdt_ops;
> +	wdd->min_timeout = 1;
> +	wdd->max_timeout = BM_WDTC_MAX(priv->wdt_freq);
> +	wdd->parent = &pdev->dev;
> +
> +	watchdog_set_drvdata(wdd, priv);
> +
> +	/*
> +	 * If 'timeout-sec' unspecified in devicetree, assume a 30 second
> +	 * default, unless the max timeout is less than 30 seconds, then use
> +	 * the max instead.
> +	 */
> +	wdd->timeout = ASM9260_WDT_DEFAULT_TIMEOUT;
> +	watchdog_init_timeout(wdd, 0, &pdev->dev);
> +
> +	watchdog_set_nowayout(wdd, nowayout);
> +
> +	asm9260_wdt_get_dt_mode(priv);
> +
> +	if (priv->mode != HW_RESET)
> +		priv->irq = platform_get_irq(pdev, 0);
> +
> +	if (priv->irq > 0) {

is_valid_irq() ?

> +		/*
> +		 * Not all supported platforms specify an interrupt for the
> +		 * watchdog, so let's make it optional.
> +		 */
> +		ret = devm_request_irq(&pdev->dev, priv->irq,
> +				asm9260_wdt_irq, IRQF_SHARED,
> +				pdev->name, priv);
> +		if (ret < 0)
> +			dev_err(&pdev->dev, "failed to request IRQ\n");

Why is this error ignored ?

> +	}
> +
> +	ret = watchdog_register_device(wdd);
> +	if (ret)
> +		goto clk_off;
> +
> +	platform_set_drvdata(pdev, priv);
> +
> +	priv->restart_handler.notifier_call = asm9260_restart_handler;
> +	priv->restart_handler.priority = 128;
> +	ret = register_restart_handler(&priv->restart_handler);
> +	if (ret)
> +		dev_err(&pdev->dev, "cannot register restart handler\n");

dev_err and abort, or dev_warn.

> +
> +	dev_info(&pdev->dev, "Watchdog enabled (timeout=%d sec, nowayout=%d)",
> +		wdd->timeout, nowayout);
> +	return 0;
> +
> +clk_off:
> +	clk_disable_unprepare(priv->clk);
> +	clk_disable_unprepare(priv->clk_ahb);

What if prepare_enable failed ?
Also, while some if not most implementations of clk_disable check
for clk==NULL and/or IS_ERR(clk), not all do.

> +	return ret;
> +}
> +
> +static void asm9260_wdt_shutdown(struct platform_device *pdev)
> +{
> +	struct asm9260_wdt_priv *priv = platform_get_drvdata(pdev);
> +
> +	asm9260_wdt_disable(&priv->wdd);
> +}
> +
> +static int __exit asm9260_wdt_remove(struct platform_device *pdev)
> +{
> +	struct asm9260_wdt_priv *priv = platform_get_drvdata(pdev);
> +
> +	asm9260_wdt_shutdown(pdev);
> +
> +	unregister_restart_handler(&priv->restart_handler);
> +	clk_disable_unprepare(priv->clk);
> +	clk_disable_unprepare(priv->clk_ahb);
> +
> +	return 0;
> +}
> +
> +static const struct of_device_id asm9260_wdt_of_match[] = {
> +	{ .compatible = "alphascale,asm9260-wdt"},
> +	{},
> +};
> +MODULE_DEVICE_TABLE(of, asm9260_wdt_of_match);
> +
> +static struct platform_driver asm9260_wdt_driver = {
> +	.driver = {
> +		.name = "asm9260-wdt",
> +		.owner = THIS_MODULE,
> +		.of_match_table	= asm9260_wdt_of_match,
> +	},
> +	.probe = asm9260_wdt_probe,
> +	.remove = asm9260_wdt_remove,
> +	.shutdown = asm9260_wdt_shutdown,
> +};
> +module_platform_driver(asm9260_wdt_driver);
> +
> +MODULE_DESCRIPTION("asm9260 WatchDog Timer Driver");
> +MODULE_AUTHOR("Oleksij Rempel <linux@rempel-privat.de>");
> +MODULE_LICENSE("GPL");
> -- 
> 2.5.0
> 
> --
> 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

  parent reply	other threads:[~2015-10-27  2:31 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-09-18  9:23 [PATCH] watchdog: add Alphascale asm9260-wdt driver Oleksij Rempel
2015-10-20  7:49 ` Oleksij Rempel
2015-10-26 12:31   ` Guenter Roeck
2015-10-27  2:31 ` Guenter Roeck [this message]
2015-10-29  7:20   ` Oleksij Rempel
2015-11-05  9:06   ` [PATCH v2] " Oleksij Rempel
2015-11-05  9:06     ` [PATCH v2] watchdog: " Oleksij Rempel
2015-11-05 16:32       ` Guenter Roeck
2015-11-05 20:43       ` Rob Herring
2015-11-23  7:11         ` Oleksij Rempel
2015-11-23  7:26           ` Guenter Roeck
2015-11-23  9:51             ` Oleksij Rempel
2015-11-23 15:47               ` Guenter Roeck
2015-11-24 21:40             ` [PATCH v3 0/2] " Oleksij Rempel
2015-11-24 21:40               ` [PATCH v3 1/2] watchdog: " Oleksij Rempel
2015-11-25  3:22                 ` Guenter Roeck
2015-11-24 21:40               ` [PATCH v3 2/2] DT: watchdog: add Alphascale asm9260 watchdog binding documentation Oleksij Rempel
2015-11-25 20:06                 ` Rob Herring
2015-11-25 19:33             ` [PATCH v4 0/2] add Alphascale asm9260-wdt driver Oleksij Rempel
2015-11-25 19:33               ` [PATCH v4 1/2] watchdog: " Oleksij Rempel
2015-11-30 16:10                 ` Guenter Roeck
2015-11-30 16:10                 ` Guenter Roeck
2015-11-25 19:33               ` [PATCH v4 2/2] DT: watchdog: add Alphascale asm9260 watchdog binding documentation Oleksij Rempel
2015-12-28 21:49               ` [PATCH v4 0/2] add Alphascale asm9260-wdt driver Wim Van Sebroeck

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=20151027023132.GA1270@roeck-us.net \
    --to=linux@roeck-us.net \
    --cc=linux-watchdog@vger.kernel.org \
    --cc=linux@rempel-privat.de \
    --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