Re: [PATCH v2] pinctrl: amd: Don't access irq_data's hwirq member directly

[Andy Shevchenko <andy.shevchenko@gmail.com>]

Fri, Aug 22, 2025 at 08:33:55AM +0200, Linus Walleij kirjoitti:
> On Fri, Aug 22, 2025 at 2:30 AM Mario Limonciello <superm1@kernel.org> wrote:
> > On 8/21/25 5:16 PM, Linus Walleij wrote:
> > > On Thu, Aug 21, 2025 at 4:49 PM Mario Limonciello (AMD)
> > > <superm1@kernel.org> wrote:
> > >
> > >> There is an irqd_to_hwirq() intended to get the hwirq number. Switch
> > >> all use to it.
> > >>
> > >> Suggested-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> > >> Signed-off-by: Mario Limonciello (AMD) <superm1@kernel.org>
> > >
> > > Hm where is this supposed to be applied...
> > > My tree?
> > >
> > > It doesn't currently apply to devel.
> > >
> > > Yours,
> > > Linus Walleij
> >
> > It was based off the commit that you picked up earlier this week.  It
> > changes lines in that based upon Andy's feedback.
> 
> Aha the single line pm_pr_debug() change. I applied that in
> fixes because it was just a oneliner and came with the other patch
> which was a fix.
> 
> But this multiline non-regression can't go into fixes.

True, but fixes may go to the next if they are needed for dependent changes.
That's why there are three ways how to handle this:
- merge fixes to devel (That's how Mark Brown does, for example)
- merge next -rcX to devel (That's how Greg KH does, for example)
- cherry-pick fixes to devel (That's how I had done, when I was PDx86 maintainer).

- (4th one) wait till next release and proceed as usual (sometimes done by some
  maintainers depending on the case at hand)

I believe we already had this discussion once or twice in the past for the
similar cases.

> I will move both of these over to -next since neither is
> a technical regression.

-- 
With Best Regards,
Andy Shevchenko