Linux Power Management development
 help / color / mirror / Atom feed
From: Patrick Titiano <ptitiano@baylibre.com>
To: magnus.damm@gmail.com, kuninori.morimoto.gx@renesas.com,
	eduardo.valentin@ti.com
Cc: bcousson@baylibre.com, linux-pm@vger.kernel.org,
	linux-sh@vger.kernel.org, Patrick Titiano <ptitiano@baylibre.com>
Subject: [RFC 0/1] thermal: rcar-thermal: add cpu cooling
Date: Fri, 28 Feb 2014 16:02:37 +0100	[thread overview]
Message-ID: <1393599758-17315-1-git-send-email-ptitiano@baylibre.com> (raw)

Hi,
This not-yet-ready-to-merge patch proposes to enable CPU thermal management on
Renesas platform. Currently these platforms only register a thermal zone,
but do not implement or register any cooling device.

With this patch:
 - CPU0 cpufreq driver is registered as a cooling device,
 - a new passive trip point is defined to trigger CPU scaling when temperature
   crosses it,
 - CPU0 cooling device and thermal zone 0 are being bound together,
 - set_trip_temp callback is implemented to facilitate experimentation and 
   platform tuning. 

Developed on a RCar-H2 platform, it is intended to support multi-platform.
It leverages CPUFreq support from the series Benoit Cousson recently shared
(http://www.spinics.net/lists/linux-sh/msg29236.html).
The passive trip point temperature is arbitrarily set to 70C, as no HW data was
available at the time this patch was developed.

Being multi-platform, this patch is expected to run smoothly in any situation:
 1/ Platform has no thermal sensor,
 2/ Platform has a thermal sensor and is DVFS capable,
 3/ Platform has a thermal sensor but is not DVFS capable.

This patch has no side-effect in case 1/ (thermal driver wouldn't be loaded).
Case 2/ is the regular case, no issue.
case 3/ is the problematic one.
Using cpufreq driver as a cooling device adds a dependency between those 2
drivers. This is taken care by deferring rcar-thermal probing until
cpufreq driver is up, but this becomes an issue when a platform does
not have DVFS support. cpufreq driver will never be probed, blocking
rcar-thermal probing.

We are still investigation what would be the best way to go, however we would
appreciate a lot any advice from the community.
Thanks!

Regards,
Patrick.

Patrick Titiano (1):
  thermal: rcar-thermal: add cpu cooling

 drivers/thermal/rcar_thermal.c | 103 ++++++++++++++++++++++++++++++++++++++---
 1 file changed, 96 insertions(+), 7 deletions(-)

-- 
1.8.3.2


             reply	other threads:[~2014-02-28 15:02 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-02-28 15:02 Patrick Titiano [this message]
2014-02-28 15:02 ` [RFC 1/1] thermal: rcar-thermal: add cpu cooling Patrick Titiano
2014-03-20  5:55   ` Zhang Rui
2014-03-20  7:57     ` Kuninori Morimoto
2014-03-20  5:55 ` [RFC 0/1] " Zhang Rui

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1393599758-17315-1-git-send-email-ptitiano@baylibre.com \
    --to=ptitiano@baylibre.com \
    --cc=bcousson@baylibre.com \
    --cc=eduardo.valentin@ti.com \
    --cc=kuninori.morimoto.gx@renesas.com \
    --cc=linux-pm@vger.kernel.org \
    --cc=linux-sh@vger.kernel.org \
    --cc=magnus.damm@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox