linux-kernel - Re: [PATCH] opp: Prepare for ->set_opp() helper to work without regulators

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]

Date:   Thu, 21 Jan 2021 16:55:36 +0530
From:   Viresh Kumar <viresh.kumar@...aro.org>
To:     Dmitry Osipenko <digetx@...il.com>
Cc:     Viresh Kumar <vireshk@...nel.org>, Nishanth Menon <nm@...com>,
        Stephen Boyd <sboyd@...nel.org>, linux-pm@...r.kernel.org,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Rafael Wysocki <rjw@...ysocki.net>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH] opp: Prepare for ->set_opp() helper to work without
 regulators

On 20-01-21, 17:50, Dmitry Osipenko wrote:
> If OPP API was meant to be thread-safe, then the
> dev_pm_opp_unregister_set_opp_helper() should unset the
> opp_table->set_opp_data under the lock since it races with
> dev_pm_opp_set_regulators().

Right, I will fix that.

> Secondly, functions like dev_pm_opp_set_rate() don't have any locks at all.

It was on purpose. It is expected that this routine specially will
only have a single caller and calls will be in sequence. This gets
called a lot and we wanted to make it as much efficient as possible.

> It should be better not to add "random" locks into the code because it
> only creates an illusion for an oblivious API user that OPP API cares
> about thread safety, IMO.
> 
> Making OPP API thread-safe will take some effort and a careful review of
> every lock will be needed.

I agree, we have kept some part out of the lock intentionally, but
every other thing which can happen in parallel is well protected.
There maybe bugs, which I am not aware of.

Another reason you see less locks is because of the way I have used
the kref thing here. That allows us to take locks for very small
section of code and not big routines.

-- 
viresh