Re: [lm-sensors] [PATCH] hwmon: (max6650) Rename the device ids to contain the hwmon suffix

[Guenter Roeck <linux@roeck-us.net>]

On Mon, Feb 10, 2014 at 06:59:55PM +0000, Laszlo Papp wrote:
> On Mon, Feb 10, 2014 at 4:53 PM,  <linux@roeck-us.net> wrote:
> > Quoting Jean Delvare <jdelvare@suse.de>:
> >
> >>
> >> That being said, going with MFD in this case seems quite overkill to
> >> me. MFD makes a lot of sense when each function has its own resources.
> >> As this isn't the case here, a single driver registering both an hwmon
> >> interface and a pinctrl interface would seem sufficient to me. But I
> >> think Guenter already discussed this in the past so I'll let him
> >> continue and decide.
> >>
> >
> > That is what I had suggested as well (though we were talking gpio
> > at the time). Laszlo didn't want to do it this way for some reason.
> > Right now I don't really have an idea what to do.
> 
> Right now I do not really have an idea what the concern here is.
> 
> I will quote you:
> 
> "Please explain, for my education, what makes you believe that I would
> object to or reject to anyone submitting such a driver."
> 
> and then the next one in the thread:
> 
> "> > Works for me. Should I apply the gpio and mfd drivers separately or in
> > > one single patch?
> >
> > s/apply/send/
> >
> Separately."
> 
> This happened about two months ago, and after two months of man work,
> several reviews from various people, while you have been *explicitly*
> included in the threads, are claiming that it is unacceptable? Do you
> see how much time waste that would be for everyone who have been
> involved.
> 
> What I currently do not understand is the point for rejecting the
> contribution that does not have API drawback, etc, if you do not
> provide anything better. You are more than welcome to rewrite my work
> once the feature works, but I guess it is very likely that you would
> not do that.
> 
> So, let me ask it: shall we continue the bike-shedding after months,
> or there is a definite decision from the maintainers? Disagreement is
> not a problem because people can move on if the maintainers actually
> make it clear what is acceptable and what not, but here that did not
> really happen. We are where we were months ago.
> 

I think I'll let Jean handle this one.

Guenter