linux-kernel - Re: Question about fixed regulator DT properties

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]

Date:	Wed, 26 Nov 2014 20:36:10 +0000
From:	Mark Brown <broonie@...nel.org>
To:	Vladimir Zapolskiy <vladimir_zapolskiy@...tor.com>
Cc:	Liam Girdwood <lgirdwood@...il.com>,
	"devicetree@...r.kernel.org" <devicetree@...r.kernel.org>,
	linux-kernel@...r.kernel.org
Subject: Re: Question about fixed regulator DT properties

On Wed, Nov 26, 2014 at 09:57:33PM +0200, Vladimir Zapolskiy wrote:

> If "regulator-boot-on" is specified and the regulator is untouched by
> bootloader/firmware, then the kernel simply enables it.

> As far as I understand the latter side-effect is exploited on quite many
> ARM boards, when there is no defined regulator consumer, but I agree
> that it looks hackish. My assumption is that probably fixed regulator

It's not just hackish, if the regulator actually needs to be on but has
no users it's actively broken since the core might decide to power it
off at any time for any reason.  Such regulators *must* be flagged as
always-on.

> logic around "regulator-boot-on" property should be changed, so that the
> kernel will not attempt to physically re-enable/enable the
> "regulator-boot-on" regulator at all, then misusage of the property
> should gone forced by necessity of finding a proper regulator consumer.

The only boards which might be able to get away with this are
non-DT/ACPI ones which aren't getting merged anyway and it is in general
going to be better to expand the set of cases where we do the disables
(since it saves power).

Download attachment "signature.asc" of type "application/pgp-signature" (474 bytes)