Re: [GIT PULL] intel-pinctrl for 6.1-1

[Andy Shevchenko <andriy.shevchenko@xxxxxxxxxxxxxxx>]

On Fri, Oct 21, 2022 at 10:02:35AM +0200, Linus Walleij wrote:
> On Thu, Oct 20, 2022 at 7:44 PM Andy Shevchenko
> <andriy.shevchenko@xxxxxxxxxxxxxxx> wrote:
> > On Thu, Oct 20, 2022 at 08:43:18PM +0300, Andy Shevchenko wrote:
> > > On Thu, Oct 20, 2022 at 07:31:23PM +0200, Bartosz Golaszewski wrote:
> > > > On Thu, Oct 20, 2022 at 4:58 PM Andy Shevchenko
> > > > <andriy.shevchenko@xxxxxxxxxxxxxxx> wrote:

...

> > > > Can you send the GPIO changes separately? This way I don't need to
> > > > pull all those pinctrl patches into the GPIO PR for the next merge
> > > > window.
> > >
> > > Some of them, but not all, if that what you wish.
> > > I.o.w. a couple of the GPIO changes must be part of pin control series.
> >
> > And I just realized that if any of new GPIO code will appear with the wrong
> > headers, there will be an inconsistent state. That said, I prefer this PR is
> > go as is.
> 
> Bart are you fine with pulling this as-is? (I am.)

Hold on a bit, I have to rebuild a whole kernel to test if I missed something
(it's already clear I missed lantiq patch, now in my branch, but obviously not
 in that tag, since I called it immutable).

> Last merge window I pulled in a big I2C cleanup (remove to return void on
> 6 million drivers) and it was fine, my diffstat looked horrible but Torvalds
> accepted it anyway I just mentioned it to him.
> 
> I think there is a way to shave off the irrelevant stuff from the diffstat,
> and I think Andy even told me how to do this but can't find/remember the
> git method used.

IIRC when you create a pull request, do it from the certain base, like
origin/master..HEAD~0 and it will drop the applied changes automatically.
But I don't remember this by heart, so need to refresh my memories as well.

-- 
With Best Regards,
Andy Shevchenko