linux-kernel - Re: [PATCH v5] watchdog: wdat_wdg: Using the existed function to check parameter timeout

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: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <7300d2b7-ab2d-ddae-199a-335bec9c5db2@roeck-us.net>
Date:   Tue, 26 Apr 2022 02:06:23 -0700
From:   Guenter Roeck <linux@...ck-us.net>
To:     Liu Xinpeng <liuxp11@...natelecom.cn>, wim@...ux-watchdog.org,
        tzungbi@...nel.org
Cc:     linux-watchdog@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v5] watchdog: wdat_wdg: Using the existed function to
 check parameter timeout

On 4/26/22 02:00, Guenter Roeck wrote:
> On 4/26/22 01:43, Liu Xinpeng wrote:
>> Context: If max_hw_heartbeat_ms is provided, the configured maximum timeout
> 
> I am quite sure I asked you to drop "Context: " and to change "existed" in
> the subject to "existing".
> 


Also, I asked you to separate the iTCO patch from the wdat series, not this one.
Why did you send this patch not as part of the other patches for the same driver ?

Guenter

> Guenter
> 
>> is not limited by it. The limit check in this driver therefore doesn't make
>> much sense. Similar, the watchdog core ensures that minimum timeout limits
>> are met if min_hw_heartbeat_ms is set. Using watchdog_timeout_invalid()
>> makes more sense because it takes this into account.
>>
>> Signed-off-by: Liu Xinpeng <liuxp11@...natelecom.cn>
>> ---
>>   drivers/watchdog/wdat_wdt.c | 5 +++--
>>   1 file changed, 3 insertions(+), 2 deletions(-)
>>
>> diff --git a/drivers/watchdog/wdat_wdt.c b/drivers/watchdog/wdat_wdt.c
>> index 195c8c004b69..9db01d165310 100644
>> --- a/drivers/watchdog/wdat_wdt.c
>> +++ b/drivers/watchdog/wdat_wdt.c
>> @@ -55,6 +55,7 @@ MODULE_PARM_DESC(nowayout, "Watchdog cannot be stopped once started (default="
>>            __MODULE_STRING(WATCHDOG_NOWAYOUT) ")");
>>   #define WDAT_DEFAULT_TIMEOUT    30
>> +#define WDAT_MIN_TIMEOUT     1
>>   static int timeout = WDAT_DEFAULT_TIMEOUT;
>>   module_param(timeout, int, 0);
>> @@ -344,6 +345,7 @@ static int wdat_wdt_probe(struct platform_device *pdev)
>>       wdat->period = tbl->timer_period;
>>       wdat->wdd.min_hw_heartbeat_ms = wdat->period * tbl->min_count;
>>       wdat->wdd.max_hw_heartbeat_ms = wdat->period * tbl->max_count;
>> +    wdat->wdd.min_timeout = WDAT_MIN_TIMEOUT;
>>       wdat->stopped_in_sleep = tbl->flags & ACPI_WDAT_STOPPED;
>>       wdat->wdd.info = &wdat_wdt_info;
>>       wdat->wdd.ops = &wdat_wdt_ops;
>> @@ -450,8 +452,7 @@ static int wdat_wdt_probe(struct platform_device *pdev)
>>        * watchdog properly after it has opened the device. In some cases
>>        * the BIOS default is too short and causes immediate reboot.
>>        */
>> -    if (timeout * 1000 < wdat->wdd.min_hw_heartbeat_ms ||
>> -        timeout * 1000 > wdat->wdd.max_hw_heartbeat_ms) {
>> +    if (watchdog_timeout_invalid(&wdat->wdd, timeout)) {
>>           dev_warn(dev, "Invalid timeout %d given, using %d\n",
>>                timeout, WDAT_DEFAULT_TIMEOUT);
>>           timeout = WDAT_DEFAULT_TIMEOUT;
>