Re: [PATCH] Staging: comedidev.h comedi_lrange should be const struct

[Greg KH]

On Mon, Apr 10, 2017 at 05:25:49PM +0200, Arthur Brainville wrote:
> On Sun, Apr 09, 2017 at 09:02:03PM +0200, Greg KH wrote:
> > On Sun, Apr 09, 2017 at 04:28:12PM +0200, Arthur Brainville (Ybalrid) wrote:
> > > According to checkpatch.pl, comedi_lrange should be declared as `const
> > > struct` instead of `struct` in driver/staging/comedidev.h
> > >
> > > Signed-off-by: Arthur Brainville (Ybalrid) <ybalrid@xxxxxxxxxxxx>
> > > ---
> > >  drivers/staging/comedi/comedidev.h | 2 +-
> > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > >
> > > diff --git a/drivers/staging/comedi/comedidev.h b/drivers/staging/comedi/comedidev.h
> > > index 1bb9986f865e..82df090783b5 100644
> > > --- a/drivers/staging/comedi/comedidev.h
> > > +++ b/drivers/staging/comedi/comedidev.h
> > > @@ -623,7 +623,7 @@ extern const struct comedi_lrange range_unknown;
> > >   * There may also be a flag that indicates the minimum and maximum are merely
> > >   * scale factors for an unknown, external reference.
> > >   */
> > > -struct comedi_lrange {
> > > +const struct comedi_lrange {
> > >  	int length;
> > >  	struct comedi_krange range[];
> > >  };
> > 
> > Huh?  Please explain how this change is correct.
> > 
> > greg k-h
> 
> First of all, thanks for taking the time to reply!
> Well, sorry if I did something wrong, I'm a newbie here, so let me
> explain:
> 
> When checking a bunch of files with /script/checkpatch.pl -f on the
> staging directory, it gave me this output:
> 
> > WARNING: struct comedi_lrange should normally be const
> > #626: FILE: drivers/staging/comedi/comedidev.h:626:
> > +struct comedi_lrange {
> > 
> > total: 0 errors, 1 warnings, 6 checks, 1043 lines checked
> > 
> > NOTE: For some of the reported defects, checkpatch may be able to
> >      mechanically convert to the typical style using --fix or
> >      --fix-inplace.
> 
> For this specific header.
> 
> Declaring it as const struct fixes that coding style warning, and
> doesn't break the build but, AFAIK isn't actually meaningfull since it's
> declaring a struct, but not any variable of type "comedi_lrange"
> at this point. And it built fine.
> I was watching an old talk (from you) about how you can send a patch to
> correct this kind of little things, so I tried.
> 
> Looking further into it, it seems that this change makes GCC trigger the
> following warning in any file that #include it :
> > In file included from drivers/staging/comedi//comedi_usb.h:24:0,
> >                 from drivers/staging/comedi//comedi_usb.c:21:
> >                 drivers/staging/comedi//comedidev.h:629:1: warning:
> >                 useless type qualifier in empty declaration
> >                  };
> 
> So, probably not good?

You should always test-build your patches, and they can never add new
kernel warnings to the build, that's not allowed at all.  So please do
that next time, it should have shown you that this was not a good
change.

And as you have found out, checkpatch.pl is a dumb tool, you still have
to think when you use it.  And if you know C, you will know that the
change you made makes no sense at all :)

thanks,

greg k-h