lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <a7ffe026-a48b-9ffd-20d3-7d25caba120d@roeck-us.net>
Date:   Mon, 27 Feb 2023 06:50:44 -0800
From:   Guenter Roeck <linux@...ck-us.net>
To:     Xingyu Wu <xingyu.wu@...rfivetech.com>
Cc:     linux-riscv@...ts.infradead.org, devicetree@...r.kernel.org,
        linux-watchdog@...r.kernel.org,
        Wim Van Sebroeck <wim@...ux-watchdog.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
        Rob Herring <robh+dt@...nel.org>,
        Paul Walmsley <paul.walmsley@...ive.com>,
        Palmer Dabbelt <palmer@...belt.com>,
        Albert Ou <aou@...s.berkeley.edu>,
        Philipp Zabel <p.zabel@...gutronix.de>,
        Samin Guo <samin.guo@...rfivetech.com>,
        linux-kernel@...r.kernel.org, Conor Dooley <conor@...nel.org>
Subject: Re: [PATCH v3 2/2] drivers: watchdog: Add StarFive Watchdog driver

On 2/26/23 22:45, Xingyu Wu wrote:
> On 2023/2/27 14:36, Guenter Roeck wrote:
>> On 2/26/23 22:26, Xingyu Wu wrote:
>>> On 2023/2/24 23:18, Guenter Roeck wrote:
>>>> On 2/23/23 23:42, Xingyu Wu wrote:
>>>>> On 2023/2/24 2:23, Guenter Roeck wrote:
>>>>>> On Mon, Feb 20, 2023 at 04:19:26PM +0800, Xingyu Wu wrote:
>>>>>>> [...]
>>>>>>> +
>>>>>>> +    wdt->wdt_device.min_timeout = 1;
>>>>>>> +    wdt->wdt_device.max_timeout = starfive_wdt_max_timeout(wdt);
>>>>>>
>>>>>>       wdt->wdt_device.timeout = STARFIVE_WDT_DEFAULT_TIME;
>>>>>>
>>>>>> should be set here. Otherwise the warning below would always be seen
>>>>>> if the module parameter is not set.
>>>>>>
>>>>>>> +
>>>>>>> +    watchdog_set_drvdata(&wdt->wdt_device, wdt);
>>>>>>> +
>>>>>>> +    /*
>>>>>>> +     * see if we can actually set the requested heartbeat,
>>>>>>> +     * and if not, try the default value.
>>>>>>> +     */
>>>>>>> +    watchdog_init_timeout(&wdt->wdt_device, heartbeat, dev);
>>>>>>> +    if (wdt->wdt_device.timeout == 0 ||
>>>>>>
>>>>>> If wdt->wdt_device.timeout is pre-initialized, it will never be 0 here.
>>>>>>
>>>>>>> +        wdt->wdt_device.timeout > wdt->wdt_device.max_timeout) {
>>>>>>
>>>>>> That won't happen because watchdog_init_timeout() validates it and does
>>>>>> not update the value if it is out of range.
>>>>>>
>>>>>>> +        dev_warn(dev, "heartbeat value out of range, default %d used\n",
>>>>>>> +             STARFIVE_WDT_DEFAULT_TIME);
>>>>>>> +        wdt->wdt_device.timeout = STARFIVE_WDT_DEFAULT_TIME;
>>>>>>
>>>>>> And this is then unnecessary. wdt->wdt_device.timeout will always be
>>>>>> valid if it was pre-initialized.
>>>>>
>>>>> It is changed to be this at beginning of the driver:
>>>>>
>>>>> static int heartbeat = STARFIVE_WDT_DEFAULT_TIME;
>>>>>
>>>>
>>>> No, this is wrong. The static variable should be set to 0 to indicate
>>>> "use default".
>>>>
>>>>> and it is changed to be this here:
>>>>>
>>>>> ret = watchdog_init_timeout(&wdt->wdt_device, heartbeat, dev);
>>>>> if (ret)
>>>>>       return ret;
>>>>>
>>>>> Would that be better?
>>>>>
>>>>
>>>> No, it is worse, because it would not instantiate the watchdog at all
>>>> if a bad heartbeat is provided.
>>>>
>>>
>>> So instantiate the watchdog with hearbeat first. And if this wrong, use default timeout.
>>> :
>>> if (watchdog_init_timeout(&wdt->wdt_device, heartbeat, dev))
>>>      wdt->wdt_device.timeout = STARFIVE_WDT_DEFAULT_TIME;
>>>
>>
>> I am kind of lost why you have to make it that complicated.
>> Just pre-initialize wdt->wdt_device.timeout like all the other drivers do,
>> and as I had suggested earlier.
>>
> 
> So you mean just use :
> wdt->wdt_device.timeout = STARFIVE_WDT_DEFAULT_TIME;
> to initialize watchdog directly?
> 

Yes, as I had suggested before, before calling watchdog_init_timeout().

Guenter

> Best regards,
> Xingyu Wu

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ