On Mon, 21 Mar 2022, Christophe Leroy wrote:
> Le 21/03/2022 à 09:50, Finn Thain a écrit :
> > Hi Christophe,
> >
> > On Mon, 21 Mar 2022, Finn Thain wrote:
> >
> >> On Mon, 21 Mar 2022, Christophe Leroy wrote:
> >>
> >>> Le 19/03/2022 à 08:20, Finn Thain a écrit :
> >>>> drivers/macintosh/via-pmu.c:897:12: warning: 'pmu_battery_proc_show' defined but not used [-Wunused-function]
> >>>> static int pmu_battery_proc_show(struct seq_file *m, void *v)
> >>>> ^~~~~~~~~~~~~~~~~~~~~
> >>>> drivers/macintosh/via-pmu.c:871:12: warning: 'pmu_irqstats_proc_show' defined but not used [-Wunused-function]
> >>>> static int pmu_irqstats_proc_show(struct seq_file *m, void *v)
> >>>> ^~~~~~~~~~~~~~~~~~~~~~
> >>>> drivers/macintosh/via-pmu.c:860:12: warning: 'pmu_info_proc_show' defined but not used [-Wunused-function]
> >>>> static int pmu_info_proc_show(struct seq_file *m, void *v)
> >>>> ^~~~~~~~~~~~~~~~~~
> >>>>
> >>>> Rearrange some code and add some #ifdefs to avoid unused code warnings
> >>>> when CONFIG_PROC_FS is disabled.
> >>>
> >>> Why not just put those three functions inside an #ifdef CONFIG_PROC_FS ?
> >>>
> >>
> >> You'd get a warning about the prototypes ("declared static but never
> >> defined"). Rather than add an ifdef around the prototypes as well, I
> >> just reordered things a little.
> >
> > Oops, I was forgetting that I also added an ifdef around the new
> > prototype.
> >
> > The simplest solution is probably the patch below, as it better exploits
> > the stubbed-out proc_* API in include/linux/proc_fs.h.
> >
> > Was this what you had in mind?
>
> Yes that's exactly what I had in mind.
>
OK, I'll submit it formally.
Thanks for your review.