[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20111206215936.GE14154@pengutronix.de>
Date: Tue, 6 Dec 2011 22:59:36 +0100
From: Wolfram Sang <w.sang@...gutronix.de>
To: Wolfgang Denk <wd@...x.de>
Cc: Heiko Schocher <hs@...x.de>, linux-watchdog@...r.kernel.org,
devicetree-discuss@...ts.ozlabs.org, linux-kernel@...r.kernel.org,
Stefan Roese <sr@...x.de>
Subject: Re: [PATCH] RFC, watchdog: add generic wdt driver API
Hi Wolfgang,
> > right, this additional API needs new drivers for all the watchdogs? Also, I
> > didn't fully get what you are missing from the combination of the current API
> > and additional userspace-handling? (check [1] for one example, not really
...
> But as Heikow wrote; this is a RFC patch, and the main purpose is to
> try to find out if it has a chance to go into mainline - several
> earlier attempts (which go back as far as 2002 - see for example [1],
> [2]) all failed with similar arguments like yours: "It's too complex,
> we already have watchdog support, nobody needs this". Funny enough
> that we have a number of customers who consider the existing wdt
> support unsufficient for their use cases. We've been using it on all
> kinds on PPC systems, and now on ARM as well.
I never said the features are useless. I also didn't say you should do
everything in userspace. I asked because I wanted to know what exactly are the
things you are missing from what is possible today. So, this could maybe be
added to the current watchdog frameworks. Adding a new framework (with lots of
issues) and requiring a new set of drivers won't cut it.
Regards,
Wolfram
--
Pengutronix e.K. | Wolfram Sang |
Industrial Linux Solutions | http://www.pengutronix.de/ |
Download attachment "signature.asc" of type "application/pgp-signature" (199 bytes)
Powered by blists - more mailing lists