linux-watchdog.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Christophe ROULLIER <christophe.roullier@st.com>
To: Guenter Roeck <linux@roeck-us.net>,
	"wim@linux-watchdog.org" <wim@linux-watchdog.org>,
	"mcoquelin.stm32@gmail.com" <mcoquelin.stm32@gmail.com>,
	Alexandre TORGUE <alexandre.torgue@st.com>
Cc: "linux-watchdog@vger.kernel.org" <linux-watchdog@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"linux-stm32@st-md-mailman.stormreply.com" 
	<linux-stm32@st-md-mailman.stormreply.com>,
	"linux-arm-kernel@lists.infradead.org" 
	<linux-arm-kernel@lists.infradead.org>
Subject: Re: [PATCH 1/1] drivers: watchdog: stm32_iwdg: set WDOG_HW_RUNNING at probe
Date: Thu, 21 Nov 2019 13:45:21 +0000	[thread overview]
Message-ID: <550566ff-31bd-1f40-63d1-67b013f8be75@st.com> (raw)
In-Reply-To: <ce10681c-8fa3-0fa6-3509-376a2f37aec9@roeck-us.net>

Hi Guenter,

On 11/21/19 10:53 AM, Guenter Roeck wrote:
> On 11/21/19 12:28 AM, Christophe Roullier wrote:
>> If the watchdog hardware is already enabled during the boot process,
>> when the Linux watchdog driver loads, it should reset the watchdog and
>> tell the watchdog framework. As a result, ping can be generated from
>> the watchdog framework (if CONFIG_WATCHDOG_HANDLE_BOOT_ENABLED is set),
>> until the userspace watchdog daemon takes over control
>>
>
> This is not what the code is doing. It sets the WDOG_HW_RUNNING flag
> unconditionally, no matter if the watchdog is already running or not.
> It also changes the semantic of the rest of the code, as well as
> functionality. The code in start_timeout no longer waits, and the ping
> code explicitly (re-)enables the watchdog.
>
> If you want an option to start the watchdog at probe time 
> unconditionally,
> please add a module parameter to do it. Otherwise you'll need to check if
> it is indeed enabled before setting WDOG_HW_RUNNING, and in that case it
> should not be necessary to re-enable it. It should also not be necessary
> to split the start function.

With our IWDG IP, there is no way to read values from hardware (if it is 
running or not)

We are in same case of intel-mid_wdt.c driver, we have Uboot which can

leaves the watchdog running. So we need to force the kicking of our 
watchdog.

Thanks

Christophe

>
> Thanks,
> Guenter
>
>> Fixes:4332d113c66a ("watchdog: Add STM32 IWDG driver")
>> Signed-off-by: Christophe Roullier <christophe.roullier@st.com>
>> ---
>>   drivers/watchdog/stm32_iwdg.c | 57 ++++++++++++++++++++++++-----------
>>   1 file changed, 40 insertions(+), 17 deletions(-)
>>
>> diff --git a/drivers/watchdog/stm32_iwdg.c 
>> b/drivers/watchdog/stm32_iwdg.c
>> index a3a329011a06..2b3be3b1c15b 100644
>> --- a/drivers/watchdog/stm32_iwdg.c
>> +++ b/drivers/watchdog/stm32_iwdg.c
>> @@ -87,8 +87,23 @@ static inline void reg_write(void __iomem *base, 
>> u32 reg, u32 val)
>>   static int stm32_iwdg_start(struct watchdog_device *wdd)
>>   {
>>       struct stm32_iwdg *wdt = watchdog_get_drvdata(wdd);
>> -    u32 tout, presc, iwdg_rlr, iwdg_pr, iwdg_sr;
>> -    int ret;
>> +
>> +    dev_dbg(wdd->parent, "%s\n", __func__);
>> +
>> +    /*  Start the watchdog */
>> +    reg_write(wdt->regs, IWDG_KR, KR_KEY_ENABLE);
>> +
>> +    /* reload watchdog */
>> +    reg_write(wdt->regs, IWDG_KR, KR_KEY_RELOAD);
>> +
>> +    set_bit(WDOG_HW_RUNNING, &wdd->status);
>> +    return 0;
>> +}
>> +
>> +static int stm32_iwdg_setprescaler(struct watchdog_device *wdd)
>> +{
>> +    struct stm32_iwdg *wdt = watchdog_get_drvdata(wdd);
>> +    u32 tout, presc, iwdg_rlr, iwdg_pr;
>>         dev_dbg(wdd->parent, "%s\n", __func__);
>>   @@ -108,19 +123,6 @@ static int stm32_iwdg_start(struct 
>> watchdog_device *wdd)
>>       /* set prescaler & reload registers */
>>       reg_write(wdt->regs, IWDG_PR, iwdg_pr);
>>       reg_write(wdt->regs, IWDG_RLR, iwdg_rlr);
>> -    reg_write(wdt->regs, IWDG_KR, KR_KEY_ENABLE);
>> -
>> -    /* wait for the registers to be updated (max 100ms) */
>> -    ret = readl_relaxed_poll_timeout(wdt->regs + IWDG_SR, iwdg_sr,
>> -                     !(iwdg_sr & (SR_PVU | SR_RVU)),
>> -                     SLEEP_US, TIMEOUT_US);
>> -    if (ret) {
>> -        dev_err(wdd->parent, "Fail to set prescaler, reload regs\n");
>> -        return ret;
>> -    }
>> -
>> -    /* reload watchdog */
>> -    reg_write(wdt->regs, IWDG_KR, KR_KEY_RELOAD);
>>         return 0;
>>   }
>> @@ -131,6 +133,9 @@ static int stm32_iwdg_ping(struct watchdog_device 
>> *wdd)
>>         dev_dbg(wdd->parent, "%s\n", __func__);
>>   +    /*  Start the watchdog */
>> +    reg_write(wdt->regs, IWDG_KR, KR_KEY_ENABLE);
>> +
>>       /* reload watchdog */
>>       reg_write(wdt->regs, IWDG_KR, KR_KEY_RELOAD);
>>   @@ -140,12 +145,21 @@ static int stm32_iwdg_ping(struct 
>> watchdog_device *wdd)
>>   static int stm32_iwdg_set_timeout(struct watchdog_device *wdd,
>>                     unsigned int timeout)
>>   {
>> +    int ret;
>> +
>>       dev_dbg(wdd->parent, "%s timeout: %d sec\n", __func__, timeout);
>>         wdd->timeout = timeout;
>>   -    if (watchdog_active(wdd))
>> -        return stm32_iwdg_start(wdd);
>> +    if (watchdog_active(wdd)) {
>> +        ret = stm32_iwdg_setprescaler(wdd);
>> +        if (ret) {
>> +            dev_err(wdd->parent, "failed to set prescaler\n");
>> +            return ret;
>> +        } else {
>> +            return stm32_iwdg_start(wdd);
>> +        }
>> +    }
>>         return 0;
>>   }
>> @@ -262,12 +276,21 @@ static int stm32_iwdg_probe(struct 
>> platform_device *pdev)
>>       watchdog_set_nowayout(wdd, WATCHDOG_NOWAYOUT);
>>       watchdog_init_timeout(wdd, 0, dev);
>>   +    /* Make sure the watchdog is serviced */
>> +    set_bit(WDOG_HW_RUNNING, &wdd->status);
>> +
>>       ret = devm_watchdog_register_device(dev, wdd);
>>       if (ret)
>>           return ret;
>>         platform_set_drvdata(pdev, wdt);
>>   +    ret = stm32_iwdg_setprescaler(wdd);
>> +    if (ret) {
>> +        dev_err(dev, "failed to set prescaler\n");
>> +        return ret;
>> +    }
>> +
>>       return 0;
>>   }
>>
>

  reply	other threads:[~2019-11-21 13:45 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-11-21  8:28 [PATCH 0/1] stm32_iwdg: set WDOG_HW_RUNNING at probe Christophe Roullier
2019-11-21  8:28 ` [PATCH 1/1] drivers: watchdog: " Christophe Roullier
2019-11-21  9:53   ` Guenter Roeck
2019-11-21 13:45     ` Christophe ROULLIER [this message]
2019-11-21 14:21       ` 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=550566ff-31bd-1f40-63d1-67b013f8be75@st.com \
    --to=christophe.roullier@st.com \
    --cc=alexandre.torgue@st.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-stm32@st-md-mailman.stormreply.com \
    --cc=linux-watchdog@vger.kernel.org \
    --cc=linux@roeck-us.net \
    --cc=mcoquelin.stm32@gmail.com \
    --cc=wim@linux-watchdog.org \
    /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 NNTP newsgroup(s).