Re: [PATCH v4 1/2] of: Rename "poweroff-source" property to "system-power-controller"

[Johan Hovold <johan@kernel.org>]

On Mon, Nov 24, 2014 at 10:24:56AM +0000, Lee Jones wrote:
> On Fri, 21 Nov 2014, Johan Hovold wrote:
> > On Thu, Nov 13, 2014 at 01:34:58PM +0000, Auto Configured wrote:
> > > From: Romain Perier <romain.perier@gmail.com>
> > > 
> > > It reverts commit a4b4e0461ec5 ("of: Add standard property for poweroff capability").
> > > As discussed on the mailing list, it makes more sense to rename back to the
> > > old established property name, without the vendor prefix. Problem being that
> > > the word "source" usually tends to be used for inputs and that is out of control
> > > of the OS. The poweroff capability is an output which simply turns the
> > > system-power off. Also, this property might be used by drivers which power-off
> > > the system and power back on subsequent RTC alarms. This seems to suggest to
> > > remove "poweroff" from the property name and to choose "system-power-controller"
> > > as the more generic name. This patchs adds the required renaming changes and
> > > defines an helper function which is compatible with both properties, the old one
> > > which was only used by tps65910 and the new one without vendor-prefix.
> > 
> > Now this is a bit of a mess.
> > 
> > There's a commit in the mfd tree, 25f833c1171d ("mfd: tps65910: Convert
> > ti,system-power-controller DT property to poweroff-source"), which
> > breaks all dts using tps65910 since these are never updated to the now
> > retracted property name ("poweroff-source").
> 
> My word!
> 
> Romain, what conversation on the MLs are you talking about?

I think Romain is referring to this thread:

	https://lkml.org/lkml/2014/10/23/161

> > This one should simply be reverted ASAP.
> 
> No need to revert, I can just remove the patch from the MFD tree.

Ok, good. Then this is limited to the regulator tree, and we could
proceed as I outlined below.

> > > Signed-off-by: Romain Perier <romain.perier@gmail.com>
> > > ---
> > >  .../bindings/power/{poweroff.txt => power-controller.txt}        | 0
> > >  .../devicetree/bindings/regulator/act8865-regulator.txt          | 4 ++--
> > >  drivers/mfd/tps65910.c                                           | 9 ++++++++-
> > >  drivers/regulator/act8865-regulator.c                            | 2 +-
> > >  include/linux/of.h                                               | 6 +++---
> > >  5 files changed, 14 insertions(+), 7 deletions(-)
> > 
> > Romain, what tree is this patch against? The changes to the tps65910
> > driver appears not to even be in the regulator tree, yet you asked Mark
> > to merge this. And the MFD maintainer is not even on CC.
> > 
> > Let's fix the breakage by reverting the offending commit in mfd. Then the
> > new standard name can be introduced in regulator alone (e.g. this patch
> > without the tps65910 bits) as nothing outside of regulator should be
> > using the new power-off feature (or binding) for act8865. Then other
> > drivers and dts can be converted to use the new property name (while
> > retaining backwards compatibility) for 3.20.
> > 
> > [ We should probably also consider adding an "of_device_is_" prefix to
> > the helper name for consistency. ]

Romain, care to resend this patch without the tps65910 chunks?

You should also fix the commit message, which claims to define a "helper
function which is compatible with both properties", something which was
no longer the case.
 
Thanks,
Johan