Re: [PATCH] iio: chemical: scd30: remove kernel.h header

[Jonathan Cameron <jonathan.cameron@oss.qualcomm.com>]

On Sun, 12 Jul 2026 13:47:24 -0500
"Maxwell Doose" <m32285159@gmail.com> wrote:

> On Sun Jul 12, 2026 at 11:52 AM CDT
> Joshua Crofts <joshua.crofts1@gmail.com> wrote:
> 
> > On Sun, 12 Jul 2026 11:42:25 -0500
> > Maxwell Doose <m32285159@gmail.com> wrote:
> >  
> >> kernel.h is no longer recommended for usage because it can cause longer
> >> build times and pull in transient dependencies. In addition, it's not
> >> being used by the SCD30 driver. Remove it.
> >> 
> >> Signed-off-by: Maxwell Doose <m32285159@gmail.com>
> >> ---
> >>  drivers/iio/chemical/scd30_core.c | 1 -
> >>  1 file changed, 1 deletion(-)
> >> 
> >> diff --git a/drivers/iio/chemical/scd30_core.c b/drivers/iio/chemical/scd30_core.c
> >> index f85cdd8bd84f..40bbf26a667b 100644
> >> --- a/drivers/iio/chemical/scd30_core.c
> >> +++ b/drivers/iio/chemical/scd30_core.c
> >> @@ -23,7 +23,6 @@
> >>  #include <linux/interrupt.h>
> >>  #include <linux/irqreturn.h>
> >>  #include <linux/jiffies.h>
> >> -#include <linux/kernel.h>
> >>  #include <linux/module.h>
> >>  #include <linux/mutex.h>
> >>  #include <linux/regulator/consumer.h>  
> >
> > Since you're doing IWYU stuff, you should also remove errno.h and
> > add err.h instead, as the code uses IS_ERR() etc. and err.h includes
> > errno.h
> >  
> 
> I tried to set up iwyu-tool to no avail, but I can send a v2 with that.
> 
> > Also consider grouping the iio headers and asm headers separately.
> > (separate patch though). The include list looks overwhelming!  
> 
> Indeed it is but we try our best. Good thing is the more we remove
> unneeded headers the better it gets.
> 
As a side note, (and I haven't checked it)
Sashiko had some interesting other stuff noticed whilst I reviewed this:
https://sashiko.dev/#/patchset/20260712164225.358852-1-m32285159%40gmail.com

If anyone has time to take a look the CMD_TEMP_OFFSET / 10 indeed looks
obviously wrong. + I moved that code without even noticing whilst using
guard() in this driver :(

Jonathan