Re: [PATCH 3/4] mfd: ti-lmu: Remove LM3532 backlight driver references

[Jacek Anaszewski]

On 4/3/19 9:57 AM, Lee Jones wrote:
> On Mon, 25 Mar 2019, Jacek Anaszewski wrote:
>
> > On 3/25/19 8:53 AM, Lee Jones wrote:
> > > On Sat, 23 Mar 2019, Jacek Anaszewski wrote:
> > >
> > > > Hi Lee,
> > > >
> > > > Can we have your ack for this going via LED tree, please?
> > >
> > > Patch looks okay.
> > >
> > > You can take it through the LED, but if you do I will need you to send
> > > me a pull-request to a minimised immutable branch please.
> > >
> > > If you cannot do this, I can apply the set and provide the same to
> > > you.
> > >
> > > If you choose the former:
> > >
> > >     Acked-for-MFD-by: Lee Jones <lee.jones@xxxxxxxxxx>
> > >
> > > Please let me know what you decide
> >
> > I've been exposing integration branches in the past, but after Linus'
> > message [0] I have my doubts now. I wonder if it wouldn't make more
> > sense if I just took the patches, and you'd cherry-pick them only in
> > case such a need occurs. This way we would avoid this whole merge
> > noise, which in an optimistic and very plausible case will not be needed
> > at all.
> >
> > [0] https://lkml.org/lkml/2017/4/19/1104
>
> That email is 2 years old, and does not seem relevant to what we're
> trying to achieve.  I've only ever had issues when *not* creating
> immutable branches for these, cross subsystem scenarios.  The
> shared branches I create are always minimalist and never change.
>
> I'm happy to take the patches and create a suitable pull-request for
> you if you are uncomfortable with the process.  I just need your Ack
> to do so.  Up to you.

I don't have any problem with the process. The clear gain of cherry
picking is more linear history. And the branch can be always created
when such a need occurs in linux-next.

That being said, I will send you a pull request once we sort out
the problem with obtaining a reference to the backlight node.

--
Best regards,
Jacek Anaszewski