linux-kernel - Re: [RFC/PATCH 0/5] DVFS in the OPP core

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: <7af27aa5-3ff6-d4ee-1dc0-bac797c2bdbc@codeaurora.org>
Date:   Thu, 7 Feb 2019 12:27:12 +0530
From:   Rajendra Nayak <rnayak@...eaurora.org>
To:     Stephen Boyd <swboyd@...omium.org>, linux-kernel@...r.kernel.org
Cc:     linux-arm-msm@...r.kernel.org, linux-pm@...r.kernel.org,
        linux-serial@...r.kernel.org, linux-spi@...r.kernel.org,
        Ulf Hansson <ulf.hansson@...aro.org>,
        Viresh Kumar <viresh.kumar@...aro.org>,
        Doug Anderson <dianders@...omium.org>
Subject: Re: [RFC/PATCH 0/5] DVFS in the OPP core

> 3) How do we handle devices that already have power-domains specified in
> DT? The opp binding for required-opps doesn't let us specify the power
> domain to target, instead it assumes that whatever power domain is
> attached to a device is the one that OPP needs to use to change the
> genpd performance state. Do we need a
> dev_pm_opp_set_required_opps_name() or something to be explicit about
> this? Can we have some way for the power domain that required-opps
> correspond to be expressed in the OPP tables themselves?

I was converting a few more drivers to use the proposed approach in this
RFC, in order to identify all outstanding issues we need to deal with,
and specifically for UFS, I end up with this exact scenario where UFS already
has an existing power domain (gdsc) and I need to add another one (rpmhpd) for
setting the performance state.

If I use dev_pm_opp_of_add_table() to add the opp table from DT, the opp
layer assumes its the same device on which it can do a dev_pm_genpd_set_performance_state()
with, however the device that's actually associated with the pm_domain when we
have multiple power domains is infact the one (dummy) that we create when
the driver makes a call to dev_pm_domain_attach_by_name/id().

Any thoughts on whats a good way to handle this?

-- 
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member
of Code Aurora Forum, hosted by The Linux Foundation