linux-kernel - Re: [RFC 3/4] power_supply: Introduce charger control interface

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

Message-ID: <20150309145517.GB950@earth>
Date:	Mon, 9 Mar 2015 15:55:17 +0100
From:	Sebastian Reichel <sre@...nel.org>
To:	"Tc, Jenny" <jenny.tc@...el.com>
Cc:	"linux-pm@...r.kernel.org" <linux-pm@...r.kernel.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	Anton Vorontsov <anton.vorontsov@...aro.org>,
	David Woodhouse <dwmw2@...radead.org>,
	"jonghwa3.lee@...sung.com" <jonghwa3.lee@...sung.com>,
	"myungjoo.ham@...il.com" <myungjoo.ham@...il.com>,
	"Pallala, Ramakrishna" <ramakrishna.pallala@...el.com>
Subject: Re: [RFC 3/4] power_supply: Introduce charger control interface

Hi,

On Mon, Mar 09, 2015 at 12:47:18PM +0000, Tc, Jenny wrote:
> > > +struct power_supply_charger {
> > > +	int (*get_property)(struct power_supply_charger *psyc,
> > > +			    enum psy_charger_control_property pspc,
> > > +			    union power_supply_propval *val);
> > 
> > The charging framework can simply call the same get_property
> > as used by sysfs. This is already done by all kind of drivers.
> 
> The idea is to separate power supply properties from power supply
> charger properties. Existing power supply properties exposes a generic
> property of a power supply. But the properties introduced above, is used
> to control charging.  But I agree, if the charger properties are moved to
> enum power_supply_property{ }, the existing set_property()/get_property()
> calls can be used

I think making them part of power_supply_property and re-using
existing functions is sensible.

> > > +	int (*set_property)(struct power_supply_charger *psyc,
> > > +			    enum psy_charger_control_property pspc,
> > > +			    const union power_supply_propval *val);
> > 
> > I guess this is needed for values, which are supposed to be
> > writable by the kernel / charging framework, but non-writable
> > by the sysfs. I suggest to add set_property_kernel() instead
> > (and make the above properties part of enum power_supply_property)
> 
> If properties are moved to enum power_supply_property {}, then it's possible
> to reuse the set_property() call. property_is_writeable() can be used to block
> user space  write access.

Right.

> > > +};
> > > +
> > >  struct power_supply {
> > >  	const char *name;
> > >  	enum power_supply_type type;
> > > @@ -200,6 +226,8 @@ struct power_supply {
> > >  	void (*external_power_changed)(struct power_supply *psy);
> > >  	void (*set_charged)(struct power_supply *psy);
> > >
> > > +	struct power_supply_charger *psy_charger;
> > 
> > Why is this a pointer?
> 
> This is introduced to access charger properties using power supply
> object.

The question was why you choose (1) over (2), considering that the
struct contains only two pointers.

(1) struct power_supply_charger *psy_charger;
(2) struct power_supply_charger psy_charger;

> If the properties can be accessed using existing
> set_property/get_property(), then this is not really needed

Right, so don't worry about my comment :)

-- Sebastian

Download attachment "signature.asc" of type "application/pgp-signature" (820 bytes)