On Saturday 14 December 2013, Jonas Jensen wrote:
> >
> > Conceptually it might be cleaner to write a separate driver, for example
> > in drivers/power/restart, than plugging the functionality into
> > the watchdog driver, at least if you don't want it in architecture
> > or platform code. The xgene restart driver is a good example.
> >
>
> Moving it to a new power/restart driver seems reasonable to me, I'm
> really just looking for a home for it, I would have preferred a
> hardware with reset in a separate register though.
>
> I now plan to remove it for v5 and submit a separate driver.
I still think it's better to leave it in the watchdog driver: This
driver attaches to the watchdog device, and adding a separate driver
for reset means we have to trick the core Linux code into attaching
two drivers to one device in some way, or the device tree has to fake
a second device that doesn't exist in hardware but is only needed
because of Linux implementation details.
Arnd
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
