linux-kernel - Re: [PATCH 7/7] regulator: core: Remove loop disabling supplies in regulator_force

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 for Android: free password hash cracker in your pocket

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

Message-ID: <20181120162504.GJ3894@sirena.org.uk>
Date:   Tue, 20 Nov 2018 16:25:04 +0000
From:   Mark Brown <broonie@...nel.org>
To:     Doug Anderson <dianders@...omium.org>
Cc:     Bjorn Andersson <bjorn.andersson@...aro.org>,
        Evan Green <evgreen@...omium.org>,
        Stephen Boyd <swboyd@...omium.org>, digetx@...il.com,
        ryandcase@...omium.org, David Collins <collinsd@...eaurora.org>,
        linux-arm-msm <linux-arm-msm@...r.kernel.org>,
        Liam Girdwood <lgirdwood@...il.com>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 7/7] regulator: core: Remove loop disabling supplies in
 regulator_force_disable()

On Tue, Nov 20, 2018 at 08:04:57AM -0800, Doug Anderson wrote:

> In general it's hard for me to reason about how the system in general
> should behave after regulator_force_disable() is called.  Is it
> basically expected that the system will panic soon after?
> Specifically other consumers of the same regulator will think it's on
> but it won't actually be on.  What should happen if one of those other
> consumers calls disable/enable?  Should the regulator turn back on?
> ...or is the regulator permanently off until the system reboots?

If something has been forced off the system is in very serious trouble
and had a cricial safety problem, though the fact that there's error
handling code that did the force disable might mean that there's some
ability to recover the situation - for example, this might be part of
thermal management or something like charger management.  Other drivers
do get notified if something gets forced off so a well written system
will ensure that other users of a regulator that can get force disabled
will have handling for this as should userspace.  We don't have any such
full systems in mainline, though - it is a really uncommon case.

The usage in the Adreno drivers just looks to be another completely
out of expectation regulator API usage in the QC code.  I do wish there
were a way to flag API calls as needing review :(

Download attachment "signature.asc" of type "application/pgp-signature" (489 bytes)