On Mon, Sep 20, 2021 at 2:31 PM Pali Rohár <pali@xxxxxxxxxx> wrote: > On Monday 20 September 2021 14:14:16 Arnd Bergmann wrote: > > From: Arnd Bergmann <arnd@xxxxxxxx> > > > > When procfs is disabled > > ... then the i8k_init_procfs function should not be called as the > purpose of I8K code / config option is to export functionality over > procfs. So when procfs is disabled then this i8k is noop. > > Patch which do not allow compilation of I8K when procfs is not enabled > is pending here: > > https://lore.kernel.org/linux-hwmon/20210910071921.16777-1-rdunlap@xxxxxxxxxxxxx/ > > Ideally please test or review it. As you are not the first one who > spotted -Werror problems with i8k and tried to workaround it. Ok, I'm now using that version in my randconfig tree, it looks sensible and it addresses another problem. I'll let you know if something else comes up with that patch applied, but I'm sure it fixes the issue I reported. Feel free to add Reported-by: Arnd Bergmann <arnd@xxxxxxxx> Reviewed-by: Arnd Bergmann <arnd@xxxxxxxx> to Randy's patch if you like. > > the code produces a warning > > for an unused variable: > > > > drivers/hwmon/dell-smm-hwmon.c: In function 'i8k_init_procfs': > > drivers/hwmon/dell-smm-hwmon.c:624:31: error: unused variable 'data' [-Werror=unused-variable] > > 624 | struct dell_smm_data *data = dev_get_drvdata(dev); > > | ^~~~ > > I'm starting to hate this -Werror decision... but seems that we have to > deal with it and together cleanup code as much as possible. Oh, I totally would have sent you my patch without the -Werror change as well ;-) Arnd
