linux-kernel - Re: [PATCH 2/2] regulator: core: always use enable

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: <20210603102741.GA4257@sirena.org.uk>
Date:   Thu, 3 Jun 2021 11:27:41 +0100
From:   Mark Brown <broonie@...nel.org>
To:     Dmitry Baryshkov <dmitry.baryshkov@...aro.org>
Cc:     Liam Girdwood <lgirdwood@...il.com>,
        Michał Mirosław <mirq-linux@...e.qmqm.pl>,
        open list <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH 2/2] regulator: core: always use enable_delay when
 enabling regulators

On Wed, Jun 02, 2021 at 04:14:35PM +0300, Dmitry Baryshkov wrote:
> On Thu, 20 May 2021 at 14:58, Mark Brown <broonie@...nel.org> wrote:

> > On Thu, May 20, 2021 at 01:12:24AM +0300, Dmitry Baryshkov wrote:
> > > Some regulators (e.g. fixed) do not have .enable callback per se, but
> > > use supply regulator and enable_delay. Do not return early from
> > > _regulator_do_enable in such cases, so that enable_delay is properly
> > > handled.

> > This doesn't seem like the right fix - if we didn't actually do anything
> > then we don't need to add a delay.  We should only be doing this if some
> > parent regulator changed state.

> I have implemented this, but then it becomes too fragile. If the
> parent gets enabled for whatever reason just few us ago, the whole
> delay would be skipped.

This seems like a solvable problem, we can track what parents are doing
or teach parents about their children for example.

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