All of lore.kernel.org
 help / color / mirror / Atom feed
* watchdoug group ping
@ 2016-04-20  7:10 fixed-term.Oleksij.Rempel
  2016-04-21 14:52 ` Guenter Roeck
  0 siblings, 1 reply; 3+ messages in thread
From: fixed-term.Oleksij.Rempel @ 2016-04-20  7:10 UTC (permalink / raw)
  To: Guenter Roeck, linux-watchdog, Dirk.Behme

Hallo all,

my target is to provide pretimout support for devices which do not have
this functionality.

Pre-requirement is this patchset (i contacted the author to rebase it
against latest watchdog core branch):
https://www.mail-archive.com/linux-watchdog@vger.kernel.org/msg02394.html

The idea is to use different watchdogs/timers or what ever as one group
- one virtual device.
One driver will be responsible for reseting the system, other for
generating an interrupt on pre-timout.
It can be also useful for systems which require redundant watchdogs -
two or more watchdog pinged over one /dev/watchdog

I assume i would need to create a master virtual watchdog driver to
bypass/broadcast all pings to child devices.

Do you have some ideas or suggestions?
-- 
Best regards,
Oleksij Rempel
CM-AI/ECO3
Tel.: +49(5121)49-2962

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: watchdoug group ping
  2016-04-20  7:10 watchdoug group ping fixed-term.Oleksij.Rempel
@ 2016-04-21 14:52 ` Guenter Roeck
  2016-04-21 20:25   ` Oleksij Rempel
  0 siblings, 1 reply; 3+ messages in thread
From: Guenter Roeck @ 2016-04-21 14:52 UTC (permalink / raw)
  To: fixed-term.Oleksij.Rempel, linux-watchdog, Dirk.Behme

On 04/20/2016 12:10 AM, fixed-term.Oleksij.Rempel wrote:
> Hallo all,
>
> my target is to provide pretimout support for devices which do not have
> this functionality.
>
> Pre-requirement is this patchset (i contacted the author to rebase it
> against latest watchdog core branch):
> https://www.mail-archive.com/linux-watchdog@vger.kernel.org/msg02394.html
>
> The idea is to use different watchdogs/timers or what ever as one group
> - one virtual device.
> One driver will be responsible for reseting the system, other for
> generating an interrupt on pre-timout.
> It can be also useful for systems which require redundant watchdogs -
> two or more watchdog pinged over one /dev/watchdog
>
> I assume i would need to create a master virtual watchdog driver to
> bypass/broadcast all pings to child devices.
>
> Do you have some ideas or suggestions?
>
Seems to add a lot of complexity. It might be much simpler to add support
for a 'soft' pretimeout in such situations, ie a timer which fires at the
correct interval and results in a reset. It might even be possible to use
the new internal heartbeat function for this purpose.

Guenter


^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: watchdoug group ping
  2016-04-21 14:52 ` Guenter Roeck
@ 2016-04-21 20:25   ` Oleksij Rempel
  0 siblings, 0 replies; 3+ messages in thread
From: Oleksij Rempel @ 2016-04-21 20:25 UTC (permalink / raw)
  To: Guenter Roeck, fixed-term.Oleksij.Rempel, linux-watchdog,
	Dirk.Behme


[-- Attachment #1.1: Type: text/plain, Size: 1342 bytes --]

Am 21.04.2016 um 16:52 schrieb Guenter Roeck:
> On 04/20/2016 12:10 AM, fixed-term.Oleksij.Rempel wrote:
>> Hallo all,
>>
>> my target is to provide pretimout support for devices which do not have
>> this functionality.
>>
>> Pre-requirement is this patchset (i contacted the author to rebase it
>> against latest watchdog core branch):
>> https://www.mail-archive.com/linux-watchdog@vger.kernel.org/msg02394.html
>>
>> The idea is to use different watchdogs/timers or what ever as one group
>> - one virtual device.
>> One driver will be responsible for reseting the system, other for
>> generating an interrupt on pre-timout.
>> It can be also useful for systems which require redundant watchdogs -
>> two or more watchdog pinged over one /dev/watchdog
>>
>> I assume i would need to create a master virtual watchdog driver to
>> bypass/broadcast all pings to child devices.
>>
>> Do you have some ideas or suggestions?
>>
> Seems to add a lot of complexity. It might be much simpler to add support
> for a 'soft' pretimeout in such situations, ie a timer which fires at the
> correct interval and results in a reset. It might even be possible to use
> the new internal heartbeat function for this purpose.

yes, you have right. it is simple and cleaner generic way.
i like it :)


-- 
Regards,
Oleksij


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 213 bytes --]

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2016-04-21 20:25 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2016-04-20  7:10 watchdoug group ping fixed-term.Oleksij.Rempel
2016-04-21 14:52 ` Guenter Roeck
2016-04-21 20:25   ` Oleksij Rempel

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.