linux-kernel - Re: [PATCH] power: supply: Allow charger manager can be built as a module

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <CADBw62oMztoidXGuv3YZ1Na9ouRAHfvuHc2yP30BccPh22443w@mail.gmail.com>
Date:   Fri, 28 Feb 2020 13:37:34 +0800
From:   Baolin Wang <baolin.wang7@...il.com>
To:     Saravana Kannan <saravanak@...gle.com>
Cc:     Orson Zhai <orsonzhai@...il.com>, sre@...nel.org,
        Linux PM <linux-pm@...r.kernel.org>,
        LKML <linux-kernel@...r.kernel.org>,
        Lyra Zhang <zhang.lyra@...il.com>,
        Android Kernel Team <kernel-team@...roid.com>
Subject: Re: [PATCH] power: supply: Allow charger manager can be built as a module

On Fri, Feb 28, 2020 at 10:54 AM Saravana Kannan <saravanak@...gle.com> wrote:
>
> On Thu, Feb 27, 2020 at 8:06 AM Orson Zhai <orsonzhai@...il.com> wrote:
> >
> > Hi Sebastian and other guys here,
> >
> > On Tue, Feb 11, 2020 at 4:51 PM Baolin Wang <baolin.wang7@...il.com> wrote:
> > >
> > > Allow charger manager can be built as a module like other charger
> > > drivers.
> > >
> > What do you think about this patch?
> > We want to set charger-manager as module in our project for new Android devices.
> >
> > -Orson
> >
> > > Signed-off-by: Baolin Wang <baolin.wang7@...il.com>
> > > ---
> > >  drivers/power/supply/Kconfig          |    2 +-
> > >  include/linux/power/charger-manager.h |    7 +------
> > >  2 files changed, 2 insertions(+), 7 deletions(-)
> > >
> > > diff --git a/drivers/power/supply/Kconfig b/drivers/power/supply/Kconfig
> > > index 9a5591a..195bc04 100644
> > > --- a/drivers/power/supply/Kconfig
> > > +++ b/drivers/power/supply/Kconfig
> > > @@ -480,7 +480,7 @@ config CHARGER_GPIO
> > >           called gpio-charger.
> > >
> > >  config CHARGER_MANAGER
> > > -       bool "Battery charger manager for multiple chargers"
> > > +       tristate "Battery charger manager for multiple chargers"
> > >         depends on REGULATOR
> > >         select EXTCON
> > >         help
> > > diff --git a/include/linux/power/charger-manager.h b/include/linux/power/charger-manager.h
> > > index ad19e68..40493b2 100644
> > > --- a/include/linux/power/charger-manager.h
> > > +++ b/include/linux/power/charger-manager.h
> > > @@ -248,11 +248,6 @@ struct charger_manager {
> > >         u64 charging_end_time;
> > >  };
> > >
> > > -#ifdef CONFIG_CHARGER_MANAGER
> > >  extern void cm_notify_event(struct power_supply *psy,
> > > -                               enum cm_event_types type, char *msg);
> > > -#else
> > > -static inline void cm_notify_event(struct power_supply *psy,
> > > -                               enum cm_event_types type, char *msg) { }
> > > -#endif
> > > +                           enum cm_event_types type, char *msg);
> > >  #endif /* _CHARGER_MANAGER_H */
>
> You are breaking the kernel if CONFIG_CHARGER_MANAGER is disabled. Why

No, now no users in mainline have called cm_notify_event() function,
and if some charger drivers want to use the charger manager, they
should select CHARGER_MANAGER, like:
config CHARGER_SC2731
        tristate "Spreadtrum SC2731 charger driver"
        depends on MFD_SC27XX_PMIC || COMPILE_TEST
        select CHARGER_MANAGER

> not simple change the #ifdef to
> #if IS_ENABLED(CONFIG_CHARGER_MANAGER)
> ?

Yes, this also can work, and either way is okay for me. Sebastian,
what do you prefer? Thanks.