Linux-PM Archive on lore.kernel.org
 help / color / Atom feed
From: Andrzej Pietrasiewicz <andrzej.p@collabora.com>
To: linux-pm@vger.kernel.org
Cc: Zhang Rui <rui.zhang@intel.com>,
	"Rafael J . Wysocki" <rjw@rjwysocki.net>,
	Len Brown <lenb@kernel.org>, Jiri Pirko <jiri@mellanox.com>,
	Ido Schimmel <idosch@mellanox.com>,
	"David S . Miller" <davem@davemloft.net>,
	Peter Kaestle <peter@piie.net>,
	Darren Hart <dvhart@infradead.org>,
	Andy Shevchenko <andy@infradead.org>,
	Support Opensource <support.opensource@diasemi.com>,
	Daniel Lezcano <daniel.lezcano@linaro.org>,
	Amit Kucheria <amit.kucheria@verdurent.com>,
	Shawn Guo <shawnguo@kernel.org>,
	Sascha Hauer <s.hauer@pengutronix.de>,
	Pengutronix Kernel Team <kernel@pengutronix.de>,
	Fabio Estevam <festevam@gmail.com>,
	NXP Linux Team <linux-imx@nxp.com>,
	Allison Randal <allison@lohutok.net>,
	Enrico Weigelt <info@metux.net>,
	Gayatri Kammela <gayatri.kammela@intel.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	linux-acpi@vger.kernel.org, netdev@vger.kernel.org,
	platform-driver-x86@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org, kernel@collabora.com,
	Andrzej Pietrasiewicz <andrzej.p@collabora.com>
Subject: [RFC 0/8] Stop monitoring disabled devices
Date: Tue,  7 Apr 2020 19:49:18 +0200
Message-ID: <20200407174926.23971-1-andrzej.p@collabora.com> (raw)

The current kernel behavior is to keep polling the thermal zone devices
regardless of their current mode. This is not desired, as all such "disabled"
devices are meant to be handled by userspace, so polling them makes no sense.

There was an attempt to solve this issue:

https://lkml.org/lkml/2018/2/26/498

and it ultimately has not succeeded:

https://lkml.org/lkml/2018/2/27/910

This is a new attempt addressing all the relevant drivers, and I have
identified them with:

$ git grep "thermal_zone_device_ops" | grep "= {" | cut -f1 -d: | sort | uniq

The idea is to modify thermal_zone_device_update() and monitor_thermal_zone()
in such a way that they stop polling a disabled device. To do decide what to
do they should call ->get_mode() operation of the specialized thermal zone
device in question (e.g. drivers/acpi/thermal.c's). But here comes problem:
sometimes a thermal zone device must be initially disabled and becomes enabled
only after its sensors appear on the system. If such thermal zone's
->get_mode() /* in the context of thermal_zone_device_update() or
monitor_thermal_zone() */ is called _before_ the sensors are available, it will
be reported as "disabled" and consequently polling it will be ceased. This is
a change in behavior from userspace's perspective.

To solve the above described problem I want to introduce the third mode of a
thermal_zone_device: initial. The idea is that when the device is in its
initial mode, then its polling will be handled as it is now. This is a good
thing: should the temperature be just about hitting the critical treshnold
early during the boot process, it might be too late if we wait for the
userspace to run to save the system from overheating. The initial mode should
be reported in sysfs as "enabled" to keep the userspace interface intact.
From the initial mode there will be two possible transitions: to enabled or
disabled mode, but there will be no transition back to initial. If the
transition is from initial to enabled, then keep polling. If the transition is
from initial to disabled, then stop polling. If the transition is from enabled
to disabled, then stop polling. The transition from disabled to enabled must
be handled in a special way: there must be a mandatory call to
monitor_thermal_zone(), otherwise the polling will not start. If this
transition is triggeted from sysfs, then it can be easily handled at the
thermal framework level. However, if drivers call their own ->set_mode()
operation then they must also call "monitor_thermal_zone()" afterwards.
The latter being a sensible thing anyway, so perhaps all/most of the drivers
in question do. The plan for implementation is this:

- ensure ALL users use symbolic enum names (THERMAL_DEVICE_DISABLED,
THERMAL_DEVICE_ENABLED) for thermal device mode rather than the numeric
values of enum thermal_device_mode elements
- add THERMAL_DEVICE_INITIAL to the said enum making its value 0 (so that
kzalloc() results in the initial state)
- modify thermal zone device's mode_show() (thermal framework level) so that
it reports "enabled" for THERMAL_DEVICE_INITIAL
- modify thermal zone device's mode_store() (thermal framework level) so that
it calls monitor_thermal_zone() upon mode change
- modify ALL thermal drivers so that their code is prepared to return
THERMAL_DEVICE_INITIAL before they call thermal_zone_device_register(); when
the invocation of the latter completes then polling is expected to be started
- verify ALL drivers which call their own ->set_mode() to ensure they do call
monitor_thermal_zone() afterwards
- modify thermal_zone_device_update() and monitor_thermal_zone() so that they
cancel polling for disabled thermal zone devices (but not for those in
THERMAL_DEVICE_INITIAL mode)

This RFC series does all the above steps in more or less that order.

I kindly ask for comments/suggestions/improvements.

Rebased onto v5.6.

Andrzej Pietrasiewicz (8):
  thermal: int3400_thermal: Statically initialize
    .get_mode()/.set_mode() ops
  thermal: Properly handle mode values in .set_mode()
  thermal: Store thermal mode in a dedicated enum
  thermal: core: Introduce THERMAL_DEVICE_INITIAL
  thermal: core: Monitor thermal zone after mode change
  thermal: Set initial state to THERMAL_DEVICE_INITIAL
  thermal: of: Monitor thermal zone after enabling it
  thermal: Stop polling DISABLED thermal devices

 drivers/acpi/thermal.c                        | 28 +++++-----
 .../ethernet/mellanox/mlxsw/core_thermal.c    | 11 +++-
 drivers/platform/x86/acerhdf.c                | 17 ++++--
 drivers/thermal/da9062-thermal.c              |  2 +-
 drivers/thermal/imx_thermal.c                 |  5 +-
 .../intel/int340x_thermal/int3400_thermal.c   | 24 ++++-----
 .../thermal/intel/intel_quark_dts_thermal.c   |  6 ++-
 drivers/thermal/of-thermal.c                  |  9 +++-
 drivers/thermal/thermal_core.c                | 52 ++++++++++++++++++-
 drivers/thermal/thermal_core.h                |  2 +
 drivers/thermal/thermal_sysfs.c               | 12 +++--
 include/linux/thermal.h                       |  3 +-
 12 files changed, 123 insertions(+), 48 deletions(-)

-- 
2.17.1


             reply index

Thread overview: 137+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-04-07 17:49 Andrzej Pietrasiewicz [this message]
2020-04-07 17:49 ` [RFC 1/8] thermal: int3400_thermal: Statically initialize .get_mode()/.set_mode() ops Andrzej Pietrasiewicz
     [not found]   ` <CGME20200415091204eucas1p1983548c2db52d8d0c2a5367034ec80dd@eucas1p1.samsung.com>
2020-04-15  9:12     ` Bartlomiej Zolnierkiewicz
2020-04-07 17:49 ` [RFC 2/8] thermal: Properly handle mode values in .set_mode() Andrzej Pietrasiewicz
     [not found]   ` <CGME20200415100543eucas1p24e24293da39844ca8791db86af5365a7@eucas1p2.samsung.com>
2020-04-15 10:05     ` Bartlomiej Zolnierkiewicz
2020-04-07 17:49 ` [RFC 3/8] thermal: Store thermal mode in a dedicated enum Andrzej Pietrasiewicz
     [not found]   ` <CGME20200415100944eucas1p1dcfeb784e790ca2fc3417fd2797e3f5d@eucas1p1.samsung.com>
2020-04-15 10:09     ` Bartlomiej Zolnierkiewicz
2020-04-07 17:49 ` [RFC 4/8] thermal: core: Introduce THERMAL_DEVICE_INITIAL Andrzej Pietrasiewicz
2020-04-07 17:49 ` [RFC 5/8] thermal: core: Monitor thermal zone after mode change Andrzej Pietrasiewicz
2020-04-07 17:49 ` [RFC 6/8] thermal: Set initial state to THERMAL_DEVICE_INITIAL Andrzej Pietrasiewicz
2020-04-07 17:49 ` [RFC 7/8] thermal: of: Monitor thermal zone after enabling it Andrzej Pietrasiewicz
     [not found]   ` <CGME20200415101829eucas1p18e21324d3c39926fffc6c8b5dc52f206@eucas1p1.samsung.com>
2020-04-15 10:18     ` Bartlomiej Zolnierkiewicz
2020-04-07 17:49 ` [RFC 8/8] thermal: Stop polling DISABLED thermal devices Andrzej Pietrasiewicz
2020-04-09 10:29 ` [RFC 0/8] Stop monitoring disabled devices Daniel Lezcano
2020-04-09 11:10   ` Andrzej Pietrasiewicz
     [not found]     ` <CGME20200415104010eucas1p101278e53e34a2e56dfc7c82b533a9122@eucas1p1.samsung.com>
2020-04-15 10:40       ` Bartlomiej Zolnierkiewicz
2020-04-17 16:23         ` Andrzej Pietrasiewicz
2020-04-19 11:42           ` Bartlomiej Zolnierkiewicz
2020-04-17 21:11         ` Peter Kästle
2020-04-14 18:00   ` [RFC v2 0/9] " Andrzej Pietrasiewicz
2020-04-14 18:00     ` [RFC v2 1/9] thermal: int3400_thermal: Statically initialize .get_mode()/.set_mode() ops Andrzej Pietrasiewicz
2020-04-15  9:23       ` Daniel Lezcano
2020-04-14 18:00     ` [RFC v2 2/9] thermal: Eliminate an always-false condition Andrzej Pietrasiewicz
2020-04-14 18:00     ` [RFC v2 3/9] thermal: Properly handle mode values in .set_mode() Andrzej Pietrasiewicz
2020-04-14 22:14       ` Daniel Lezcano
2020-04-14 18:01     ` [RFC v2 4/9] thermal: core: Let thermal zone device's mode be stored in its struct Andrzej Pietrasiewicz
2020-04-14 22:30       ` Daniel Lezcano
2020-04-14 18:01     ` [RFC v2 5/9] thermal: Store mode in thermal_zone_device Andrzej Pietrasiewicz
2020-04-15 10:55       ` Daniel Lezcano
2020-04-17 16:20         ` [RFC v3 0/2] Stop monitoring disabled devices Andrzej Pietrasiewicz
2020-04-17 16:20           ` [RFC v3 1/2] thermal: core: Let thermal zone device's mode be stored in its struct Andrzej Pietrasiewicz
2020-04-17 20:44             ` Andy Shevchenko
2020-04-19 11:38             ` Bartlomiej Zolnierkiewicz
2020-04-19 13:10               ` Bartlomiej Zolnierkiewicz
2020-04-20 18:17                 ` [PATCH 0/2] Stop monitoring disabled devices Andrzej Pietrasiewicz
2020-04-20 18:17                   ` [PATCH 1/2] thermal: core: Let thermal zone device's mode be stored in its struct Andrzej Pietrasiewicz
2020-04-22  7:49                     ` kbuild test robot
2020-04-22 17:14                       ` Andrzej Pietrasiewicz
2020-04-22 17:03                     ` kbuild test robot
2020-04-22 17:15                       ` Andrzej Pietrasiewicz
2020-04-20 18:17                   ` [PATCH 2/2] thermal: core: Stop polling DISABLED thermal devices Andrzej Pietrasiewicz
2020-04-22 16:12                   ` [PATCH RESEND 1/2] thermal: core: Let thermal zone device's mode be stored in its struct Andrzej Pietrasiewicz
2020-04-23 14:39                     ` Bartlomiej Zolnierkiewicz
2020-04-23 16:57                       ` [PATCH v3 0/2] Stop monitoring disabled devices Andrzej Pietrasiewicz
2020-04-23 16:57                         ` [PATCH v3 1/2] thermal: core: Let thermal zone device's mode be stored in its struct Andrzej Pietrasiewicz
2020-05-04  7:00                           ` Bartlomiej Zolnierkiewicz
2020-04-23 16:57                         ` [PATCH v3 2/2] thermal: core: Stop polling DISABLED thermal devices Andrzej Pietrasiewicz
2020-04-24  9:03                           ` Zhang, Rui
2020-04-27 14:20                             ` Zhang, Rui
2020-04-27 18:34                               ` Andrzej Pietrasiewicz
2020-04-28 13:55                                 ` Zhang, Rui
2020-04-20 11:03               ` [RFC v3 1/2] thermal: core: Let thermal zone device's mode be stored in its struct Andrzej Pietrasiewicz
2020-04-20 18:19                 ` Andrzej Pietrasiewicz
2020-05-23 21:24             ` Daniel Lezcano
2020-05-25 19:35               ` Andrzej Pietrasiewicz
2020-05-25 22:08                 ` Daniel Lezcano
2020-05-26 16:56                   ` Andrzej Pietrasiewicz
2020-05-27 13:30               ` Bartlomiej Zolnierkiewicz
2020-04-17 16:20           ` [RFC v3 2/2] thermal: core: Stop polling DISABLED thermal devices Andrzej Pietrasiewicz
2020-04-19 13:50           ` [RFC v3 1/2] thermal: core: Let thermal zone device's mode be stored in its struct Peter Kästle
2020-04-14 18:01     ` [RFC v2 6/9] thermal: Remove get_mode() method Andrzej Pietrasiewicz
2020-04-14 18:01     ` [RFC v2 7/9] thermal: core: Monitor thermal zone after mode change Andrzej Pietrasiewicz
2020-04-14 18:01     ` [RFC v2 8/9] thermal: of: Monitor thermal zone after enabling it Andrzej Pietrasiewicz
2020-04-14 18:01     ` [RFC v2 9/9] thermal: core: Stop polling DISABLED thermal devices Andrzej Pietrasiewicz
2020-04-14 21:52     ` [RFC v2 0/9] Stop monitoring disabled devices Ezequiel Garcia
     [not found] <Message-ID: <4493c0e4-51aa-3907-810c-74949ff27ca4@samsung.com>
2020-05-28 19:20 ` [PATCH v4 00/11] " Andrzej Pietrasiewicz
2020-05-28 19:20   ` [PATCH v4 01/11] acpi: thermal: Fix error handling in the register function Andrzej Pietrasiewicz
2020-05-29 14:50     ` Guenter Roeck
2020-06-24  8:16     ` Bartlomiej Zolnierkiewicz
2020-05-28 19:20   ` [PATCH v4 02/11] thermal: Store thermal mode in a dedicated enum Andrzej Pietrasiewicz
2020-05-29 14:48     ` Guenter Roeck
2020-05-29 15:13       ` Andrzej Pietrasiewicz
2020-05-29 15:34         ` Guenter Roeck
2020-06-24  9:38     ` Bartlomiej Zolnierkiewicz
2020-05-28 19:20   ` [PATCH v4 03/11] thermal: Add current mode to thermal zone device Andrzej Pietrasiewicz
2020-05-29 14:51     ` Guenter Roeck
2020-06-24  9:39     ` Bartlomiej Zolnierkiewicz
2020-05-28 19:20   ` [PATCH v4 04/11] thermal: Store device mode in struct thermal_zone_device Andrzej Pietrasiewicz
2020-06-24  9:40     ` Bartlomiej Zolnierkiewicz
2020-05-28 19:20   ` [PATCH v4 05/11] thermal: remove get_mode() operation of drivers Andrzej Pietrasiewicz
2020-06-24  9:47     ` Bartlomiej Zolnierkiewicz
2020-05-28 19:20   ` [PATCH v4 06/11] thermal: Add mode helpers Andrzej Pietrasiewicz
2020-06-24  9:49     ` Bartlomiej Zolnierkiewicz
2020-05-28 19:20   ` [PATCH v4 07/11] thermal: Use mode helpers in drivers Andrzej Pietrasiewicz
2020-06-24  9:51     ` Bartlomiej Zolnierkiewicz
2020-06-26 16:08       ` Andrzej Pietrasiewicz
2020-05-28 19:20   ` [PATCH v4 08/11] thermal: Explicitly enable non-changing thermal zone devices Andrzej Pietrasiewicz
2020-06-24 10:00     ` Bartlomiej Zolnierkiewicz
2020-05-28 19:20   ` [PATCH v4 09/11] thermal: core: Stop polling DISABLED thermal devices Andrzej Pietrasiewicz
2020-06-24 10:02     ` Bartlomiej Zolnierkiewicz
2020-05-28 19:20   ` [PATCH v4 10/11] thermal: Simplify or eliminate unnecessary set_mode() methods Andrzej Pietrasiewicz
2020-06-24 10:03     ` Bartlomiej Zolnierkiewicz
2020-05-28 19:20   ` [PATCH v4 11/11] thermal: Rename set_mode() to change_mode() Andrzej Pietrasiewicz
2020-06-24 10:03     ` Bartlomiej Zolnierkiewicz
2020-06-01 11:36   ` [PATCH v4 02/11] thermal: Store thermal mode in a dedicated enum Peter Kästle
2020-06-01 11:36   ` [PATCH v4 04/11] thermal: Store device mode in struct thermal_zone_device Peter Kästle
2020-06-01 11:37   ` [PATCH v4 05/11] thermal: remove get_mode() operation of drivers Peter Kästle
2020-06-01 11:37   ` [PATCH v4 07/11] thermal: Use mode helpers in drivers Peter Kästle
2020-06-01 11:38   ` [PATCH v4 10/11] thermal: Simplify or eliminate unnecessary set_mode() methods Peter Kästle
2020-06-01 11:38   ` [PATCH v4 11/11] thermal: Rename set_mode() to change_mode() Peter Kästle
2020-06-23 14:37   ` [PATCH v4 00/11] Stop monitoring disabled devices Daniel Lezcano
2020-06-26 17:37     ` [PATCH v5 " Andrzej Pietrasiewicz
2020-06-26 17:37       ` [PATCH v5 01/11] acpi: thermal: Fix error handling in the register function Andrzej Pietrasiewicz
2020-06-29  7:19         ` Amit Kucheria
2020-06-26 17:37       ` [PATCH v5 02/11] thermal: Store thermal mode in a dedicated enum Andrzej Pietrasiewicz
2020-06-29  7:19         ` Amit Kucheria
2020-06-26 17:37       ` [PATCH v5 03/11] thermal: Add current mode to thermal zone device Andrzej Pietrasiewicz
2020-06-29  7:19         ` Amit Kucheria
2020-06-26 17:37       ` [PATCH v5 04/11] thermal: Store device mode in struct thermal_zone_device Andrzej Pietrasiewicz
2020-06-29  7:19         ` Amit Kucheria
2020-06-26 17:37       ` [PATCH v5 05/11] thermal: remove get_mode() operation of drivers Andrzej Pietrasiewicz
2020-06-29  7:20         ` Amit Kucheria
2020-06-26 17:37       ` [PATCH v5 06/11] thermal: Add mode helpers Andrzej Pietrasiewicz
2020-06-26 20:50         ` kernel test robot
2020-06-26 20:50         ` [RFC PATCH] thermal: thermal_zone_device_set_mode() can be static kernel test robot
2020-06-26 20:55         ` [PATCH v5 06/11] thermal: Add mode helpers kernel test robot
2020-06-29  7:04         ` Amit Kucheria
2020-06-29 11:16           ` [PATCH v6 00/11] Stop monitoring disabled devices Andrzej Pietrasiewicz
2020-06-29 11:16             ` [PATCH v6 01/11] acpi: thermal: Fix error handling in the register function Andrzej Pietrasiewicz
2020-06-29 11:16             ` [PATCH v6 02/11] thermal: Store thermal mode in a dedicated enum Andrzej Pietrasiewicz
2020-06-29 11:16             ` [PATCH v6 03/11] thermal: Add current mode to thermal zone device Andrzej Pietrasiewicz
2020-06-29 11:16             ` [PATCH v6 04/11] thermal: Store device mode in struct thermal_zone_device Andrzej Pietrasiewicz
2020-06-29 11:16             ` [PATCH v6 05/11] thermal: remove get_mode() operation of drivers Andrzej Pietrasiewicz
2020-06-29 11:16             ` [PATCH v6 06/11] thermal: Add mode helpers Andrzej Pietrasiewicz
2020-06-29 12:08               ` Daniel Lezcano
2020-06-29 11:16             ` [PATCH v6 07/11] thermal: Use mode helpers in drivers Andrzej Pietrasiewicz
2020-06-29 11:16             ` [PATCH v6 08/11] thermal: Explicitly enable non-changing thermal zone devices Andrzej Pietrasiewicz
2020-06-29 11:16             ` [PATCH v6 09/11] thermal: core: Stop polling DISABLED thermal devices Andrzej Pietrasiewicz
2020-06-29 11:16             ` [PATCH v6 10/11] thermal: Simplify or eliminate unnecessary set_mode() methods Andrzej Pietrasiewicz
2020-06-29 11:16             ` [PATCH v6 11/11] thermal: Rename set_mode() to change_mode() Andrzej Pietrasiewicz
2020-06-26 17:37       ` [PATCH v5 07/11] thermal: Use mode helpers in drivers Andrzej Pietrasiewicz
2020-06-26 17:37       ` [PATCH v5 08/11] thermal: Explicitly enable non-changing thermal zone devices Andrzej Pietrasiewicz
2020-06-26 17:37       ` [PATCH v5 09/11] thermal: core: Stop polling DISABLED thermal devices Andrzej Pietrasiewicz
2020-06-26 17:37       ` [PATCH v5 10/11] thermal: Simplify or eliminate unnecessary set_mode() methods Andrzej Pietrasiewicz
2020-06-26 17:37       ` [PATCH v5 11/11] thermal: Rename set_mode() to change_mode() Andrzej Pietrasiewicz
2020-06-01 10:02 ` [PATCH v4 00/11] Stop monitoring disabled devices Peter Kästle
2020-06-01 10:20   ` Andrzej Pietrasiewicz

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=20200407174926.23971-1-andrzej.p@collabora.com \
    --to=andrzej.p@collabora.com \
    --cc=allison@lohutok.net \
    --cc=amit.kucheria@verdurent.com \
    --cc=andy@infradead.org \
    --cc=daniel.lezcano@linaro.org \
    --cc=davem@davemloft.net \
    --cc=dvhart@infradead.org \
    --cc=festevam@gmail.com \
    --cc=gayatri.kammela@intel.com \
    --cc=idosch@mellanox.com \
    --cc=info@metux.net \
    --cc=jiri@mellanox.com \
    --cc=kernel@collabora.com \
    --cc=kernel@pengutronix.de \
    --cc=lenb@kernel.org \
    --cc=linux-acpi@vger.kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-imx@nxp.com \
    --cc=linux-pm@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=peter@piie.net \
    --cc=platform-driver-x86@vger.kernel.org \
    --cc=rjw@rjwysocki.net \
    --cc=rui.zhang@intel.com \
    --cc=s.hauer@pengutronix.de \
    --cc=shawnguo@kernel.org \
    --cc=support.opensource@diasemi.com \
    --cc=tglx@linutronix.de \
    /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

Linux-PM Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-pm/0 linux-pm/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-pm linux-pm/ https://lore.kernel.org/linux-pm \
		linux-pm@vger.kernel.org
	public-inbox-index linux-pm

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-pm


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git