Re: [PATCH v3] leds: is31fl319x: modernize registration

On Wed, 12 Aug 2026 14:29:41 +0100
Lee Jones <[email protected]> wrote:

> On Wed, 12 Aug 2026, Andreas Kemnade wrote:
> 
> > On Tue, 11 Aug 2026 19:51:27 +0100
> > Lee Jones <[email protected]> wrote:
> >   
> > > On Thu, 06 Aug 2026, Andreas Kemnade wrote:
> > >   
> > > > On Thu, 6 Aug 2026 11:45:25 +0100
> > > > Lee Jones <[email protected]> wrote:
> > > >     
> > > > > On Tue, 04 Aug 2026, Andreas Kemnade wrote:
> > > > >     
> > > > > > On Tue, 04 Aug 2026 13:55:14 +0000
> > > > > > [email protected] wrote:
> > > > > >       
> > > > > > > > leds: is31fl319x: modernize registration        
> > > > > > > 
> > > > > > > [Severity: Low]
> > > > > > > This isn't a bug, but should the description after the subsystem prefix be
> > > > > > > capitalized? The LED subsystem guidelines recommend formatting commit messages
> > > > > > > as "leds: <Driver>: <Capitalized description>".      
> > > > > > 
> > > > > > 
> > > > > > ~/linux$ grep -R 'leds: <Driver>: <Capitalized description>"' Documentation/
> > > > > > ~/linux$ ls Documentation/leds/
> > > > > > index.rst                  leds-lp5521.rst      leds-sc27xx.rst
> > > > > > leds-blinkm.rst            leds-lp5523.rst      leds-st1202.rst
> > > > > > leds-cht-wcove.rst         leds-lp5562.rst      ledtrig-oneshot.rst
> > > > > > leds-class-flash.rst       leds-lp55xx.rst      ledtrig-transient.rst
> > > > > > leds-class-multicolor.rst  leds-lp5812.rst      ledtrig-usbport.rst
> > > > > > leds-class.rst             leds-mlxcpld.rst     uleds.rst
> > > > > > leds-el15203000.rst        leds-mt6370-rgb.rst  well-known-leds.txt
> > > > > > leds-lm3556.rst            leds-qcom-lpg.rst
> > > > > > leds-lp3944.rst            leds-s2m-rgb.rst
> > > > > > 
> > > > > > nothing look like special guidelines for the system. I do not want to
> > > > > > see such comments without a reference.
> > > > > > Also MAINTAINERS does not show any additional policy for the LED subsystem.      
> > > > > 
> > > > > `git log --oneline -- <subsystem>` is your friend.
> > > > >     
> > > > That is looking at common practice, not guidelines, which is of course
> > > > often a good idea.
> > > > 
> > > > So Sahiko is hallucinating about non-existant guidelines which
> > > > cannot be referenced of course. Well, at least it gives quite an amount of
> > > > useful hintse.    
> > > 
> > > It's not hallucinating, I put it there. =:-)
> > > 
> > > https://github.com/sashiko-dev/sashiko/blob/main/third_party/prompts/kernel/subsystem/leds.md
> > > 
> > > Specifically:
> > > 
> > > https://github.com/sashiko-dev/sashiko/blob/main/third_party/prompts/kernel/subsystem/leds.md?plain=1#L8
> > >   
> > So, that is another repository of another project. That is not an
> > intuitive place to look at. So I would expect to see at least some pointer
> > in that direction in the first place.  
> 
> It's not common for maintainers to document all of their idiosyncrasies
> and preferences.  You've asked if the review comment is legitimate and
> I've told you that it is.
> 
It is not common to document everything. That is correct. 
I think you got me a bit wrong here.

I have not questioned the legitimacy of the review comment, I am questioning
the style of this Sashiko review which is basically: "there is documentation
but I do not tell you where it is, even if it is in unusual place (I consider
anything not in the Documentation folder of the kernel git as unusal)."

Regards,
Andreas