[U-Boot] [PATCH v2 3/5] cmd: bcb: Fix duplicated handling in two case-branches

[Sam Protsenko]

On Fri, Jul 12, 2019 at 6:49 PM Eugeniu Rosca <erosca at de.adit-jv.com> wrote:
>
> On Fri, Jul 12, 2019 at 05:01:09PM +0300, Sam Protsenko wrote:
> > > > diff --git a/cmd/bcb.c b/cmd/bcb.c
> > > > index 2bd5a744deb5..3b1c7434e287 100644
> > > > --- a/cmd/bcb.c
> > > > +++ b/cmd/bcb.c
> > > > @@ -46,9 +46,6 @@ static int bcb_is_misused(int argc, char *const argv[])
> > > >
> > > >         switch (cmd) {
> > > >         case BCB_CMD_LOAD:
> > > > -               if (argc != 3)
> > > > -                       goto err;
> > > > -               break;
> >
> > Wait, one note here... Can you please add /* Fall through */ comment
> > here? I remember that new GCC was complaining on  cases without break
> > that don't have such comment or corresponding fall-through attribute.
> > I can see that U-Boot build doesn't emit any warnings for that case,
> > but I think such comment would be useful anyway.
> >
> > > >         case BCB_CMD_FIELD_SET:
> > > >                 if (argc != 3)
> > > >                         goto err;
>
> Both U-Boot and Linux enable [-Wimplicit-fallthrough=] on "make W=1".
> Regardless of U-Boot and Linux, gcc doesn't seem to report any warning
> when two case statements are stacked on top of each other without any
> logic in between. The warning is only triggered if there is minimal
> processing done in between (e.g. a dummy printf call).
>
> If we still want the /* fallthrough */ comment, I will add it in the
> next patch revision.
>

Understood. Ok, let's not worry about this then. My R-b tag stands.

> --
> Best Regards,
> Eugeniu.