Aw: Re: [PATCH 3/3] mpc52xx/wdt: WDT uses GPT api
Albrecht Dreß
albrecht.dress at arcor.de
Wed Nov 11 19:32:22 EST 2009
Hi Grant:
O.k., thanks for your comments. If Wim doesn't have any objections to it, I will provide a merged patch. One consequence I forgot to mention is that we loose the ability to build the wdt support as module, but I don't think it's a real problem.
I think we still should keep the kernel config option enable/disable the wdt support, which would mask out the wdt code if disabled. Is that ok for you?
Thanks, Albrecht.
----- Original Nachricht ----
Von: Grant Likely <grant.likely at secretlab.ca>
An: Albrecht Dreß <albrecht.dress at arcor.de>
Datum: 10.11.2009 22:07
Betreff: Re: [PATCH 3/3] mpc52xx/wdt: WDT uses GPT api
> On Tue, Nov 10, 2009 at 1:26 PM, Albrecht Dreß <albrecht.dress at arcor.de>
> wrote:
> > Hi Grant:
> >
> > Am 10.11.09 20:59 schrieb(en) Grant Likely:
> >>
> >> On Tue, Nov 10, 2009 at 12:43 PM, Albrecht Dreß
> <albrecht.dress at arcor.de>
> >> wrote:
> >> > Use the MPC5200 GPT api for the WDT which drastically simplifies this
> >> > file.
> >> >
> >> > Signed-off-by: Albrecht Dreß <albrecht.dress at arcor.de>
> >> > ---
> >> >
> >> > drivers/watchdog/mpc5200_wdt.c | 246
> >> > +++++++++++-----------------------------
> >> > 1 files changed, 65 insertions(+), 181 deletions(-)
> >>
> >>
> >> Can the WDT functionality just be merged entirely into
> >> arch/powerpc/platforms/52xx/mpc52xx_gpt.c, eliminating the need for
> >> this file entirely? I think I'd rather have all the GPT "built in"
> >> behaviour handled by a single driver.
> >
> > I also thought about it, as it has IMHO the cleaner code, and it would
> have
> > the extra benefit that the gpt-wdt api doesn't need to be public.
> >
> > However, the reasons I hesitated to do so are:
> > - I don't want to remove a file someone else wrote (even it doesn't
> work);
>
> Shouldn't be a problem, and I'll handle the fallout if it is.
>
> > - WDT code is shifted from drivers/watchdog to
> arch/powerpc/platforms/52xx
> > which might not be the "logical" place from the directory layout's pov;
>
> There is precedence of this in the past, particularly on arch or
> platform specific hardware drivers and multifunction devices. (Heck,
> that's almost entirely what arch/powerpc/sysdev is). I'm not
> concerned.
>
> > - a file living in arch/powerpc/platforms/52xx depends upon config
> options
> > set from drivers/watchdog/Kconfig which may be confusing.
>
> I'm not concerned about this either.
>
> > You see these are more political/cosmetical questions, so I would prefer
> to
> > leave the decision to the maintainers (i.e. you and Wim). Preparing a
> fully
> > merged driver is actually a matter of minutes!
>
> Do it. I'll champion getting it in. Wim, do you have any issues with
> this?
>
> g.
>
> --
> Grant Likely, B.Sc., P.Eng.
> Secret Lab Technologies Ltd.
> _______________________________________________
> Linuxppc-dev mailing list
> Linuxppc-dev at lists.ozlabs.org
> https://lists.ozlabs.org/listinfo/linuxppc-dev
>
Jetzt NEU: Do it youself E-Cards bei Arcor.de!
Stellen Sie Ihr eigenes Unikat zusammen und machen Sie dem Empfänger eine ganz persönliche Freude!
E-Card Marke Eigenbau: HIER KLICKEN: http://www.arcor.de/rd/footer.ecard
More information about the devicetree-discuss
mailing list