linux-kernel - Re: Mess in the /sysfs: class/power

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: <201502252301.41633@pali>
Date:	Wed, 25 Feb 2015 23:01:41 +0100
From:	Pali Rohár <pali.rohar@...il.com>
To:	Sebastian Reichel <sre@...nel.org>
Cc:	Pavel Machek <pavel@....cz>,
	kernel list <linux-kernel@...r.kernel.org>,
	Greg KH <greg@...ah.com>, dbaryshkov@...il.com,
	dwmw2@...radead.org, linux-pm@...r.kernel.org
Subject: Re: Mess in the /sysfs: class/power_supply/bq2415x

On Wednesday 25 February 2015 22:44:42 Sebastian Reichel wrote:
> Hi Pavel,
> 
> On Sat, Jan 31, 2015 at 10:39:19PM +0100, Pavel Machek wrote:
> > So we have multiple values in one sysfs file:
> > 
> > root@...0:/sys/class/power_supply# cat bq24150a-0/registers
> > 0x00=0x50
> > 0x01=0xc8
> > 0x02=0x8e
> > 0x03=0x4b
> > 0x04=0x99
> > 
> > (should this go to debugfs somewhere?)
> 
> I guess this should be dropped in favour of debugfs. A
> standardized debugfs node will be exported automatically if
> the driver is converted to regmap.
> 
> > Unobvious and undocumented entries:
> > 
> > root@...0:/sys/class/power_supply# cat
> > bq24150a-0/boost_status 0
> > pavel@duo:/data/l/linux-n900$ grep -ri boost_status
> > Documentation/ pavel@duo:/data/l/linux-n900$ grep -ri
> > termination_current Documentation/
> > pavel@duo:/data/l/linux-n900$ grep -ri current_now
> > Documentation/ pavel@duo:/data/l/linux-n900$ grep -ri
> > time_to_empty_avg Documentation/
> > pavel@duo:/data/l/linux-n900$
> 
> Anton seems to have missed making Pali to write a
> Documentation patch. Pali: Can you do so, please?
> 

Some of properties are standard power_supply properties and those 
are documented in Documentation/power/power_supply_class.txt

bq2415x specific are documented only in bq2415x_charger.c source 
code. So probably we should move documentation into 
Documentation/power/bq2415x_charer.txt

But I think there are other stuff with higher priority for N900 
as moving comments/documentation...

> > And different units in different parts of tree:
> > 
> > (these are mA)
> > 
> > root@...0:/sys/class/power_supply# cat
> > bq24150a-0/current_limit 1800
> > root@...0:/sys/class/power_supply# cat
> > bq24150a-0/termination_current 100
> > 

Those are non standard power-supply properties and so unit is not 
defined.

> > (and this is in uA)
> > 
> > root@...0:/sys/class/power_supply# cat bq27200-0/current_now
> > 12138
> > 

This is power supply.

> > (I believe uA is right, so it is bq24150a that should be
> > fixed....?)
> > 

Units
~~~~~
Quoting include/linux/power_supply.h:

  All voltages, currents, charges, energies, time and 
temperatures in µV,
  µA, µAh, µWh, seconds and tenths of degree Celsius unless 
otherwise
  stated. It's driver's job to convert its raw values to units in 
which
  this class operates.

> > Ok, of course questions are: do we need to do something with
> > it? If so, what?
> 
> For consistency it should be uA, but I guess this would
> introduce a regression for Maemo people. Pali?
> 
> -- Sebastian

It looks like that in Maemo & CSSU we are using only:

/sys/class/power_supply/bq24150a-0/mode
/sys/class/power_supply/bq24150a-0/stat_pin_enable

But it does not mean that some other people do not use other 
bq2415x sysfs nodes.

-- 
Pali Rohár
pali.rohar@...il.com

Download attachment "signature.asc " of type "application/pgp-signature" (199 bytes)