Hello Greg, On Wed, 2020-03-18 at 14:08 +0100, Greg Kroah-Hartman wrote: > On Tue, Feb 25, 2020 at 10:53:01AM +0200, Matti Vaittinen wrote: > > Many devices have control registers which control some measurable > > property. Often a register contains control field so that change in > > this field causes linear change in the controlled property. It is > > not > > a rare case that user wants to give 'meaningful' control values and > > driver needs to convert them to register field values. Even more > > often user wants to 'see' the currently set value - again in > > meaningful units - and driver needs to convert the values it reads > > from register to these meaningful units. Examples of this include: > > > > - regulators, voltage/current configurations > > - power, voltage/current configurations > > - clk(?) NCOs > > > > and maybe others I can't think of right now. > > > > Provide a linear_range helper which can do conversion from user > > value > > to register value 'selector'. > > > > The idea here is stolen from regulator framework and patches > > refactoring > > the regulator helpers to use this are following. > > > > Signed-off-by: Matti Vaittinen <matti.vaittinen@xxxxxxxxxxxxxxxxx> > > --- > > > > Changes since rfc-v3: > > - Kerneldoc fixes > > - Corrected commit message typo meaningfull => meaningful > > > > drivers/base/Kconfig | 3 + > > drivers/base/Makefile | 1 + > > drivers/base/linear_ranges.c | 246 > > +++++++++++++++++++++++++++++++++++ > > Why in drivers/base/ ? > > Why not in lib/ ? I was pondering which of these would be better. I decided to do with drivers/base because - in it's current form - this is really a driver related stuff. I see it somehow in same position as regmap code - although this is just a tiny helper compared to regmap. But this also has pretty driver specific audience :) And... I must admit I like things which I know. And I have been doing driver development and "know" a few of the driver related colleagues - hence working with them is easier for me ;) Getting to know the colleagues maintaining lib is a bit scary :] Yep, I'm Finnish if you happen to wonder why getting to know people is scary xD > > > include/linux/linear_range.h | 48 +++++++ > > 4 files changed, 298 insertions(+) > > create mode 100644 drivers/base/linear_ranges.c > > create mode 100644 include/linux/linear_range.h > > > > diff --git a/drivers/base/Kconfig b/drivers/base/Kconfig > > index 5f0bc74d2409..636b6fa8e499 100644 > > --- a/drivers/base/Kconfig > > +++ b/drivers/base/Kconfig > > @@ -209,4 +209,7 @@ config GENERIC_ARCH_TOPOLOGY > > appropriate scaling, sysfs interface for reading capacity > > values at > > runtime. > > > > +config LINEAR_RANGES > > + tristate > > No help text at all??? Yes. The linear ranges has no meaning to be enabled alone. It only plays a role if it is used by some driver/subsystem. And drivers/subsystems should do select LINEAR_RANGES. So showing help in any config tool is not needed. This should actually not be visible in menuconfig or others. I think I have seen a few examples like this. Ayways, I have no obejctions to adding some text if absolutely needed. Any suggestions for a text politely saying - "please, pretend I am not here" - are welcome :) (Although, I think this really does not need help text). > > > +EXPORT_SYMBOL(linear_range_values_in_range); > > EXPORT_SYMBOL_GPL() for all of these? I have to ask... I personally have no objections towards EXPORT_SYMBOL_GPL. I guess regulator helpers and the power supply modules which use this are GPL modules. If no other (better) opinions, then I can change this in next version. Thanks for pointing it out - I didn't even think about it. > > > @@ -0,0 +1,48 @@ > > +/* SPDX-License-Identifier: GPL-2.0-or-later */ > > Are you sure about the "or later"? Again, I have to ask. No. If you want to educate me on this topic - or point a link to some nice explanation why to not use "later" - pretty please. I would like to learn :) > > > +/* Copyright (C) 2020 ROHM Semiconductors */ > > + > > +#ifndef LINEAR_RANGE_H > > +#define LINEAR_RANGE_H > > + > > +#include <linux/types.h> > > Why is this needed? I think it was the bool type which was missing without this. At least on my ARM gcc toolchain. Best Regards Matti Vaittinen
