From: Guenter Roeck <linux@roeck-us.net>
To: Mathieu Poirier <mathieu.poirier@linaro.org>, fu.wei@linaro.org
Cc: "Rob Herring" <robh+dt@kernel.org>,
"Paweł Moll" <pawel.moll@arm.com>,
"Mark Rutland" <mark.rutland@arm.com>,
"Ian Campbell" <ijc+devicetree@hellion.org.uk>,
"Kumar Gala" <galak@codeaurora.org>,
wim@iguana.be, "Jon Corbet" <corbet@lwn.net>,
"Catalin Marinas" <catalin.marinas@arm.com>,
"Will Deacon" <will.deacon@arm.com>,
Suravee.Suthikulpanit@amd.com,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
linux-watchdog@vger.kernel.org, linux-doc@vger.kernel.org,
"devicetree@vger.kernel.org" <devicetree@vger.kernel.org>,
"linux-arm-kernel@lists.infradead.org"
<linux-arm-kernel@lists.infradead.org>,
linaro-acpi@lists.linaro.org, rruigrok@codeaurora.org,
harba@codeaurora.org,
"Christopher Covington" <cov@codeaurora.org>,
timur@codeaurora.org, dyoung@redhat.com, panand@redhat.com,
graeme.gregory@linaro.org, al.stone@linaro.org,
hanjun.guo@linaro.org, jcm@redhat.com,
"Arnd Bergmann" <arnd@arndb.de>,
leo.duran@amd.com, "Sudeep Holla" <sudeep.holla@arm.com>
Subject: Re: [PATCH v10 5/5] Watchdog: ARM SBSA Generic Watchdog half timeout panic support
Date: Thu, 4 Feb 2016 08:41:55 -0800 [thread overview]
Message-ID: <56B37F53.9060002@roeck-us.net> (raw)
In-Reply-To: <CANLsYkyLBfSKFM6CCyPP+Eaua2B+n80wL2hnQVSvfUNMPd3OdQ@mail.gmail.com>
On 02/04/2016 08:32 AM, Mathieu Poirier wrote:
> On 3 February 2016 at 10:18, <fu.wei@linaro.org> wrote:
>> From: Fu Wei <fu.wei@linaro.org>
>>
>> This patch registers the WS0 interrupt routine to trigger panic,
>> when the watchdog reachs the first stage (the half timeout).
>> This function can help administrator to backup the system context
>> info by panic console output or kdump (if supported), once system
>> goes wrong (doesn't feed the watchdog in the half timeout).
>>
>> User also can skip panic by setting panic_enabled (module parameter) as 0
>>
>> Signed-off-by: Fu Wei <fu.wei@linaro.org>
>> ---
>> Documentation/watchdog/watchdog-parameters.txt | 1 +
>> drivers/watchdog/Kconfig | 10 +++++
>> drivers/watchdog/sbsa_gwdt.c | 54 +++++++++++++++++++++++---
>> 3 files changed, 60 insertions(+), 5 deletions(-)
>>
>> diff --git a/Documentation/watchdog/watchdog-parameters.txt b/Documentation/watchdog/watchdog-parameters.txt
>> index 300eb4d..31641e2 100644
>> --- a/Documentation/watchdog/watchdog-parameters.txt
>> +++ b/Documentation/watchdog/watchdog-parameters.txt
>> @@ -286,6 +286,7 @@ nowayout: Watchdog cannot be stopped once started
>> -------------------------------------------------
>> sbsa_gwdt:
>> timeout: Watchdog timeout in seconds. (default 20s)
>> +panic_enabled: Enable panic at half timeout. (default=true)
>> nowayout: Watchdog cannot be stopped once started
>> (default=kernel config parameter)
>> -------------------------------------------------
>> diff --git a/drivers/watchdog/Kconfig b/drivers/watchdog/Kconfig
>> index 4ab1b05..42adfdf 100644
>> --- a/drivers/watchdog/Kconfig
>> +++ b/drivers/watchdog/Kconfig
>> @@ -218,6 +218,16 @@ config ARM_SBSA_WATCHDOG
>> To compile this driver as module, choose M here: The module
>> will be called sbsa_gwdt.
>>
>> +config ARM_SBSA_WATCHDOG_PANIC
>> + bool "ARM SBSA Generic Watchdog triggers panic at the half timeout"
>> + depends on ARM_SBSA_WATCHDOG
>> + help
>> + ARM SBSA Generic Watchdog will trigger panic in the first signal
>> + (WS0) interrupt routine when the half timeout is reached.
>> + This function can help administrator to backup the system context
>> + info by panic console output or kdump (if supported).
>> + But user can skip panic by setting moduleparam panic_enabled as 0.
>> +
>> config ASM9260_WATCHDOG
>> tristate "Alphascale ASM9260 watchdog"
>> depends on MACH_ASM9260
>> diff --git a/drivers/watchdog/sbsa_gwdt.c b/drivers/watchdog/sbsa_gwdt.c
>> index 5a2dba3..d18cf37 100644
>> --- a/drivers/watchdog/sbsa_gwdt.c
>> +++ b/drivers/watchdog/sbsa_gwdt.c
>> @@ -16,18 +16,22 @@
>> * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
>> * GNU General Public License for more details.
>> *
>> - * This SBSA Generic watchdog driver is a single stage timeout version.
>> + * This SBSA Generic watchdog driver is a two stages version.
>> * Since this watchdog timer has two stages, and each stage is determined
>> * by WOR. So the timeout is (WOR * 2).
>> - * When first timeout is reached, WS0 is triggered, the interrupt
>> - * triggered by WS0 will be ignored, then the second watch period starts;
>> - * when second timeout is reached, then WS1 is triggered, system reset.
>> + * When the first stage(the half timeout) is reached, WS0 interrupt is
>> + * triggered, at this moment the second watch period starts;
>> + * In the WS0 interrupt routine, panic will be triggered for saving the
>> + * system context.
>> + * If the system is getting into trouble and cannot be reset by panic or
>> + * restart properly by the kdump kernel(if supported), then the second
>> + * stage (the timeout) will be reached, system will be reset by WS1.
>> *
>> * More details about the hardware specification of this device:
>> * ARM DEN0029B - Server Base System Architecture (SBSA)
>> *
>> * SBSA GWDT: |--------WOR-------WS0--------WOR-------WS1
>> - * |----------------timeout----------------reset
>> + * |--half_timeout--(panic)--half_timeout--reset
>> *
>> */
>>
>> @@ -84,6 +88,13 @@ MODULE_PARM_DESC(timeout,
>> "Watchdog timeout in seconds. (>=0, default="
>> __MODULE_STRING(DEFAULT_TIMEOUT) ")");
>>
>> +#ifdef CONFIG_ARM_SBSA_WATCHDOG_PANIC
>> +static bool panic_enabled = true;
>> +module_param(panic_enabled, bool, 0);
>> +MODULE_PARM_DESC(panic_enabled,
>> + "enable panic at half timeout. (default=true)");
>> +#endif
>> +
>> static bool nowayout = WATCHDOG_NOWAYOUT;
>> module_param(nowayout, bool, S_IRUGO);
>> MODULE_PARM_DESC(nowayout,
>> @@ -159,6 +170,16 @@ static int sbsa_gwdt_stop(struct watchdog_device *wdd)
>> return 0;
>> }
>>
>> +#ifdef CONFIG_ARM_SBSA_WATCHDOG_PANIC
>> +static irqreturn_t sbsa_gwdt_interrupt(int irq, void *dev_id)
>> +{
>> + if (panic_enabled)
>> + panic("SBSA Watchdog half timeout");
>> +
>> + return IRQ_HANDLED;
>> +}
>> +#endif
>> +
>> static struct watchdog_info sbsa_gwdt_info = {
>> .identity = "SBSA Generic Watchdog",
>> .options = WDIOF_SETTIMEOUT |
>> @@ -186,6 +207,9 @@ static int sbsa_gwdt_probe(struct platform_device *pdev)
>> struct resource *res;
>> u32 status;
>> int ret;
>> +#ifdef CONFIG_ARM_SBSA_WATCHDOG_PANIC
>> + int irq;
>> +#endif
>>
>> gwdt = devm_kzalloc(dev, sizeof(*gwdt), GFP_KERNEL);
>> if (!gwdt)
>> @@ -202,6 +226,14 @@ static int sbsa_gwdt_probe(struct platform_device *pdev)
>> if (IS_ERR(rf_base))
>> return PTR_ERR(rf_base);
>>
>> +#ifdef CONFIG_ARM_SBSA_WATCHDOG_PANIC
>> + irq = platform_get_irq(pdev, 0);
>> + if (irq < 0) {
>> + dev_err(dev, "unable to get ws0 interrupt.\n");
>> + return irq;
>> + }
>> +#endif
>> +
>
> Can't the driver revert to single stage mode if platform_get_irq()
> fails? That way the value of 'irq' can be tested throughout the
> _probe() function and the #ifdefs removed.
>
Good point. Making irq support mandatory on all systems isn't really
a good idea.
Thanks,
Guenter
> Thanks,
> Mathieu
>
>> /*
>> * Get the frequency of system counter from the cp15 interface of ARM
>> * Generic timer. We don't need to check it, because if it returns "0",
>> @@ -228,6 +260,14 @@ static int sbsa_gwdt_probe(struct platform_device *pdev)
>> dev_warn(dev, "System reset by WDT.\n");
>> wdd->bootstatus |= WDIOF_CARDRESET;
>> }
>> +#ifdef CONFIG_ARM_SBSA_WATCHDOG_PANIC
>> + ret = devm_request_irq(dev, irq, sbsa_gwdt_interrupt, 0,
>> + pdev->name, gwdt);
>> + if (ret) {
>> + dev_err(dev, "unable to request IRQ %d\n", irq);
>> + return ret;
>> + }
>> +#endif
>>
>> ret = watchdog_register_device(wdd);
>> if (ret)
>> @@ -242,6 +282,10 @@ static int sbsa_gwdt_probe(struct platform_device *pdev)
>>
>> dev_info(dev, "Initialized with %ds timeout @ %u Hz%s\n", wdd->timeout,
>> gwdt->clk, status & SBSA_GWDT_WCS_EN ? " [enabled]" : "");
>> +#ifdef CONFIG_ARM_SBSA_WATCHDOG_PANIC
>> + dev_info(dev, "Half timeout panic %s.\n",
>> + panic_enabled ? "enabled" : "disabled");
>> +#endif
>>
>> return 0;
>> }
>> --
>> 2.5.0
>>
>
next prev parent reply other threads:[~2016-02-04 16:42 UTC|newest]
Thread overview: 46+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-02-03 17:18 [PATCH v10 0/5] Watchdog: introduce ARM SBSA watchdog driver fu.wei
2016-02-03 17:18 ` [PATCH v10 1/5] Documentation: add sbsa-gwdt driver documentation fu.wei
2016-02-03 17:18 ` [PATCH v10 2/5] ARM64: add SBSA Generic Watchdog device node in foundation-v8.dts fu.wei
2016-02-03 17:18 ` [PATCH v10 3/5] ARM64: add SBSA Generic Watchdog device node in amd-seattle-soc.dtsi fu.wei
2016-02-03 17:18 ` [PATCH v10 4/5] Watchdog: introduce ARM SBSA watchdog driver fu.wei
2016-02-03 17:48 ` Timur Tabi
2016-02-03 17:57 ` Fu Wei
2016-02-03 17:58 ` Timur Tabi
2016-02-03 18:09 ` Fu Wei
2016-02-03 18:12 ` Timur Tabi
2016-02-04 16:25 ` Mathieu Poirier
2016-02-05 9:01 ` Fu Wei
2016-02-04 16:31 ` Will Deacon
2016-02-04 16:37 ` Timur Tabi
2016-02-04 16:46 ` Guenter Roeck
2016-02-05 8:52 ` Fu Wei
2016-02-03 17:18 ` [PATCH v10 5/5] Watchdog: ARM SBSA Generic Watchdog half timeout panic support fu.wei
2016-02-03 17:27 ` Timur Tabi
2016-02-03 17:49 ` Fu Wei
2016-02-03 17:53 ` Timur Tabi
2016-02-03 18:06 ` Fu Wei
2016-02-03 18:08 ` Timur Tabi
2016-02-03 18:18 ` Fu Wei
2016-02-03 18:26 ` Timur Tabi
2016-02-03 18:37 ` Fu Wei
2016-02-03 18:45 ` Timur Tabi
2016-02-03 23:00 ` Fu Wei
2016-02-04 5:17 ` Guenter Roeck
2016-02-04 13:48 ` Timur Tabi
2016-02-04 14:41 ` Guenter Roeck
2016-02-05 9:51 ` Fu Wei
2016-02-05 13:02 ` Thomas Petazzoni
2016-02-05 13:08 ` Timur Tabi
2016-02-05 13:33 ` Thomas Petazzoni
2016-02-05 14:42 ` Guenter Roeck
2016-02-05 18:21 ` Fu Wei
2016-02-05 23:54 ` Guenter Roeck
2016-02-06 18:02 ` Fu Wei
2016-02-06 18:55 ` Timur Tabi
2016-02-07 2:02 ` Fu Wei
2016-02-06 18:57 ` Guenter Roeck
2016-02-07 2:12 ` Fu Wei
2016-02-04 16:32 ` Mathieu Poirier
2016-02-04 16:41 ` Guenter Roeck [this message]
2016-02-04 16:43 ` Timur Tabi
2016-02-05 9:20 ` Fu Wei
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=56B37F53.9060002@roeck-us.net \
--to=linux@roeck-us.net \
--cc=Suravee.Suthikulpanit@amd.com \
--cc=al.stone@linaro.org \
--cc=arnd@arndb.de \
--cc=catalin.marinas@arm.com \
--cc=corbet@lwn.net \
--cc=cov@codeaurora.org \
--cc=devicetree@vger.kernel.org \
--cc=dyoung@redhat.com \
--cc=fu.wei@linaro.org \
--cc=galak@codeaurora.org \
--cc=graeme.gregory@linaro.org \
--cc=hanjun.guo@linaro.org \
--cc=harba@codeaurora.org \
--cc=ijc+devicetree@hellion.org.uk \
--cc=jcm@redhat.com \
--cc=leo.duran@amd.com \
--cc=linaro-acpi@lists.linaro.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-doc@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-watchdog@vger.kernel.org \
--cc=mark.rutland@arm.com \
--cc=mathieu.poirier@linaro.org \
--cc=panand@redhat.com \
--cc=pawel.moll@arm.com \
--cc=robh+dt@kernel.org \
--cc=rruigrok@codeaurora.org \
--cc=sudeep.holla@arm.com \
--cc=timur@codeaurora.org \
--cc=will.deacon@arm.com \
--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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).