Re: [PATCH] staging: fbtft: Use ARRAY_SIZE() to get argument count

[Julia Lawall <julia.lawall@inria.fr>]

On Mon, 31 Oct 2022, Deepak R Varma wrote:

> On Mon, Oct 31, 2022 at 12:41:40PM +0530, Deepak Varma wrote:
> > On Sat, Oct 29, 2022 at 07:34:26PM +0200, Julia Lawall wrote:
> > >
> > >
> > > On Sat, 29 Oct 2022, Deepak R Varma wrote:
> > >
> > > > On Sat, Oct 29, 2022 at 09:32:50AM +0200, Greg Kroah-Hartman wrote:
> > > > > On Fri, Oct 28, 2022 at 07:00:05PM +0530, Deepak R Varma wrote:
> > > > > > The ARRAY_SIZE(foo) macro should be preferred over sizeof operator
> > > > > > based computation such as sizeof(foo)/sizeof(foo[0]) for finding
> > > > > > number of elements in an array. Issue identified using coccicheck.
> > > > > >
> > > > > > Signed-off-by: Deepak R Varma <drv@mailo.com>
> > > > > > ---
> > > > > >  drivers/staging/fbtft/fbtft.h | 2 +-
> > > > > >  1 file changed, 1 insertion(+), 1 deletion(-)
> > > > > >
> > > > > > diff --git a/drivers/staging/fbtft/fbtft.h b/drivers/staging/fbtft/fbtft.h
> > > > > > index 2c2b5f1c1df3..5506a473be91 100644
> > > > > > --- a/drivers/staging/fbtft/fbtft.h
> > > > > > +++ b/drivers/staging/fbtft/fbtft.h
> > > > > > @@ -231,7 +231,7 @@ struct fbtft_par {
> > > > > >  	bool polarity;
> > > > > >  };
> > > > > >
> > > > > > -#define NUMARGS(...)  (sizeof((int[]){__VA_ARGS__}) / sizeof(int))
> > > > > > +#define NUMARGS(...)  ARRAY_SIZE(((int[]){ __VA_ARGS__ }))
> > > > >
> > > > > Please please please test-build your patches before sending them out.
> > > > > To not do so just wastes reviewer resources :(
> > > >
> > > > Hello Greg,
> > > > I did build the .ko files by making the driver/staging/fbtft/ path. I verified
> > > > .o and .ko files were built.
> > > >
> > > > I did a make clean just now and was again able to rebuild without any errors.
> > > > Please see the attached log file.
> > > >
> > > > Is there something wrong with the way I am firing the build?
> > >
> > > The change is in the definition of a macro.  The compiler won't help you
> > > in this case unless the macro is actually used in code that is compiled.
> > > Find the uses and check for any nearby ifdefs.  For file foo.c you can
> > > also do make foo.i to see the result of reducing ifdef and expanding
> > > macros.  Then you can see if the code you changed is actually included in
> > > the build.
> >
> > Okay. This is helpful. I understand. Looking into the file where the macro
> > expansion is reported to be failed.
>
> Hi Julia,
> I could see the macro expansions in the .i files for the fbtft-core.c and
> fb_hx8353d.c file. I am not sure why it built successfully on my x86 though. The
> error in Kerbel bot seems to be specific to ARM arch. I will try that later
> today. I am on the right track to the build error triage?
>
> Also, while reviewing the macro expansion, I saw change in the computation that
> seems odd to me. In the denominator of the expanded macro, there is a "+
> ((int)...." computation that I am not sure if is result of ARRAY_SIZE. I have
> attached the old anf the new .i file diff for your review. If you get a change
> could you help me understand why this additional computation is added to the
> denominator?

I took a look, but it's pretty complex.  You could take the code and
reorganize it so that it is more readable, and then take the definition of
the ARRAY_SIZE macro, to better see what is going on.

julia

>
> Thank you,
> ./drv
> >
> > Thank you,
> > ./drv
> >
> > >
> > > julia
> > >
> >
> >
> >
>
>
>