linux-kernel - Re: [PATCH] sbs-battery: add option to always register battery

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: <20150603135750.GB18181@earth>
Date:	Wed, 3 Jun 2015 15:57:51 +0200
From:	Sebastian Reichel <sre@...nel.org>
To:	Frans Klaver <frans.klaver@...ns.com>
Cc:	Dmitry Eremin-Solenikov <dbaryshkov@...il.com>,
	David Woodhouse <dwmw2@...radead.org>,
	linux-pm@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] sbs-battery: add option to always register battery

Hi Frans,

On Tue, Jun 02, 2015 at 03:14:43PM +0200, Frans Klaver wrote:
> Commit a22b41a31e53 ("sbs-battery: Probe should try talking to the
> device") introduced a step in probing the SBS battery, that tries to
> talk to the device before actually registering it, saying:
> 
>     this driver doesn't actually try talking to the device at probe
>     time, so if it's incorrectly configured in the device tree or
>     platform data (or if the battery has been removed from the system),
>     then probe will succeed and every access will sit there and time
>     out. The end result is a possibly laggy system that thinks it has a
>     battery but can never read status, which isn't very useful.
> 
> Which is of course reasonable. However, it is also very well possible
> for a device to boot up on wall-power and be connected to a battery
> later on. The current advice in this situation is to probe the device
> from userspace if you expect the battery to come on at some point in the
> future. The downside of this approach is that userspace needs to be
> aware of the backend of its powersupply, which is inconvenient and going
> against the point of hardware abstraction.
> 
> In some of these cases you do want to register a battery, even if none
> are attached at the moment. To facilitate this, add a configuration
> option to try to talk to the device, defaulting to y, thus keeping the
> current behavior. If unset, the battery will always be registered
> without checking the sanity of the connection.
> 
> Signed-off-by: Frans Klaver <frans.klaver@...ns.com>
> ---
> If there's a better place to arrange for this all to happen, or to make this
> more common across power supplies, I'm perfectly happy to do that work instead.
> For now this seems like the logical step to take, especially since using device
> tree was (sensibly) shot down last september [0].

While I still think, that the HW design is bad, I'm basically fine
with this change based upon your comments. I think it's better to
make this into a module parameter, though, since that moves the
decision about this feature from compilation time to module load
time. This will make it possible to use a generic kernel on your
device. Maybe something like this could be used:

module_param(force_load, bool, 0444);
MODULE_PARM_DESC(force_load,
		 "Attempts to load the driver even if the "
		 "battery is not connected");

-- Sebastian

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