linux-kernel - RE: [PATCH v1 1/2] watchdog: add global watchdog kernel module parameters structure

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

Message-ID: <5689370584614335b4f43660d80a5338@asem.it>
Date:   Tue, 9 Mar 2021 10:38:48 +0000
From:   Flavio Suligoi <f.suligoi@...m.it>
To:     Randy Dunlap <rdunlap@...radead.org>,
        Wim Van Sebroeck <wim@...ux-watchdog.org>,
        Guenter Roeck <linux@...ck-us.net>,
        Mika Westerberg <mika.westerberg@...ux.intel.com>
CC:     "linux-watchdog@...r.kernel.org" <linux-watchdog@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: RE: [PATCH v1 1/2] watchdog: add global watchdog kernel module
 parameters structure

> Hi,
> 

Hi Randy,

> On 3/8/21 3:21 AM, Flavio Suligoi wrote:
> > Different watchdog modules frequently require the same type of
> > parameters (for example: timeout, nowayout feature, start wdog on
> > module insertion, etc.).

...

> > +In this way, each driver can read these "global" parameters and then,
> > +if needed, can implement them, according to the particular hw
> > +watchdog
> 
> Please spell out "hardware" (not "hw").

Ok!

> 
> > +characteristic.
> > +
> > +Using this approach, it is possible reduce some duplicate code in the
> > +*new*
> 
>                               possible to reduce

Thanks!

...

> > + * Instead of add this kind of module parameters independently to
> > +each driver,
> 
>                  adding

Thanks!

...
> > + * if needed, implements them, according to the particular hw
> > + watchdog
> 
> s/hw/hardware/

Ok

...

> If I were doing (or using) this, I would probably want 'test_mode' and
> 'verbosity' to be unsigned int masks instead of a bool, so that there could be
> multiple types of test_mode or verbosity.

I used bool as already done in some watchdog drivers, but your suggestion
is better, thanks.

> That's something that some other subsystems do, but maybe watchdog is
> simple enough that it's not needed.
> If it is needed, then we are back to each driver doing it its own way (until this
> patch is updated).
> 
> 
> HTH. thanks.
> --
> ~Randy

Thanks Randy.

Regards,
Flavio