Hi Andy,
On Tue, Feb 4, 2020 at 8:30 PM Andy Shevchenko
<andriy.shevchenko@xxxxxxxxxxxxxxx> wrote:
> On Tue, Feb 04, 2020 at 08:26:21PM +0200, Laurent Pinchart wrote:
> > On Tue, Feb 04, 2020 at 06:39:34PM +0100, Geert Uytterhoeven wrote:
> > > On Tue, Feb 4, 2020 at 5:20 PM Andy Shevchenko wrote:
> > > > Replace with appropriate types.h.
> > >
> > > Thanks for your patch!
> > >
> > > I have only one very short question: why?
> >
> > Likewise :-) The patch itself looks fine, but the commit message is a
> > bit terse.
>
> The kernel.h for a long time being a dump of a lot of things. I started
> cleaning it up a bit. During this I notice that developers too far too lazy to
> use appropriate headers. For platform data kernel.h by definition is not
> appropriate.
Thanks, that makes perfect sense.
> Any suggestion what should I put to commit message?
I became intrigued by the one-line summary, which seemed to suggest
(according to my interpretation) that including kernel.h is a Real Bad
Thing.
But basically all you wanted to say was:
drm: shmobile: Reduce include dependencies
and:
This file doesn't need everything provided by <linux/kernel.h>.
All it needs are some types, which are provided by <linux/types.h>.
Right?
BTW, <drm/drm_mode.h> already includes <linux/types.h>, but not relying
on implicit includes is indeed a good thing.
Gr{oetje,eeting}s,
Geert
--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds
