* Re: [patch 3/5] cpuidle: add haltpoll governor
From: Marcelo Tosatti @ 2019-06-27 18:08 UTC (permalink / raw)
To: Doug Smythies
Cc: 'Paolo Bonzini', 'Radim Krcmar',
'Andrea Arcangeli', 'Rafael J. Wysocki',
'Peter Zijlstra', 'Wanpeng Li',
'Konrad Rzeszutek Wilk', 'Raslan KarimAllah',
'Boris Ostrovsky', 'Ankur Arora',
'Christian Borntraeger', linux-pm, 'kvm-devel'
In-Reply-To: <002e01d527c9$23be6e10$6b3b4a30$@net>
On Thu, Jun 20, 2019 at 05:34:46PM -0700, Doug Smythies wrote:
> Hi,
>
> I tried your patch set, but only to check
> that they didn't cause any regression for situations
> where idle state 0 (Poll) is used a lot (teo governor).
>
> They didn't (my testing was not thorough).
>
> I do not know if the below matters or not.
>
> On 2019.06.13 15:46 Marcelo Tosatti wrote:
>
> ... [snip] ...
>
> > Index: linux-2.6.git/Documentation/virtual/guest-halt-polling.txt
> > ===================================================================
> > --- /dev/null 1970-01-01 00:00:00.000000000 +0000
> > +++ linux-2.6.git/Documentation/virtual/guest-halt-polling.txt 2019-06-13 18:16:22.414262777 -0400
> > @@ -0,0 +1,79 @@
> > +Guest halt polling
> > +==================
> > +
> > +The cpuidle_haltpoll driver, with the haltpoll governor, allows
> > +the guest vcpus to poll for a specified amount of time before
> > +halting.
> > +This provides the following benefits to host side polling:
> > +
> > + 1) The POLL flag is set while polling is performed, which allows
> > + a remote vCPU to avoid sending an IPI (and the associated
> > + cost of handling the IPI) when performing a wakeup.
> ^
> |_ While applying the patches, git complains about this space character before the TAB.
>
> It also complains about a few patches with a blank line before EOF.
>
> ... Doug
Hi Doug,
Will fix those, thanks.
^ permalink raw reply
* [PATCH 35/43] docs: thermal: convert to ReST
From: Mauro Carvalho Chehab @ 2019-06-28 12:20 UTC (permalink / raw)
To: Linux Doc Mailing List
Cc: Mauro Carvalho Chehab, Mauro Carvalho Chehab, linux-kernel,
Jonathan Corbet, Kukjin Kim, Krzysztof Kozlowski, Zhang Rui,
Eduardo Valentin, Daniel Lezcano, linux-arm-kernel,
linux-samsung-soc, linux-pm, Arjan van de Ven
In-Reply-To: <cover.1561723979.git.mchehab+samsung@kernel.org>
Rename the thermal documentation files to ReST, add an
index for them and adjust in order to produce a nice html
output via the Sphinx build system.
At its new index.rst, let's add a :orphan: while this is not linked to
the main index.rst file, in order to avoid build warnings.
Signed-off-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org>
---
...pu-cooling-api.txt => cpu-cooling-api.rst} | 39 +-
.../{exynos_thermal => exynos_thermal.rst} | 47 +-
.../thermal/exynos_thermal_emulation | 53 --
.../thermal/exynos_thermal_emulation.rst | 61 +++
Documentation/thermal/index.rst | 18 +
...el_powerclamp.txt => intel_powerclamp.rst} | 177 +++----
.../{nouveau_thermal => nouveau_thermal.rst} | 54 +-
...ower_allocator.txt => power_allocator.rst} | 140 ++---
.../thermal/{sysfs-api.txt => sysfs-api.rst} | 490 ++++++++++++------
...hermal => x86_pkg_temperature_thermal.rst} | 28 +-
MAINTAINERS | 2 +-
include/linux/thermal.h | 4 +-
12 files changed, 689 insertions(+), 424 deletions(-)
rename Documentation/thermal/{cpu-cooling-api.txt => cpu-cooling-api.rst} (82%)
rename Documentation/thermal/{exynos_thermal => exynos_thermal.rst} (67%)
delete mode 100644 Documentation/thermal/exynos_thermal_emulation
create mode 100644 Documentation/thermal/exynos_thermal_emulation.rst
create mode 100644 Documentation/thermal/index.rst
rename Documentation/thermal/{intel_powerclamp.txt => intel_powerclamp.rst} (76%)
rename Documentation/thermal/{nouveau_thermal => nouveau_thermal.rst} (64%)
rename Documentation/thermal/{power_allocator.txt => power_allocator.rst} (74%)
rename Documentation/thermal/{sysfs-api.txt => sysfs-api.rst} (66%)
rename Documentation/thermal/{x86_pkg_temperature_thermal => x86_pkg_temperature_thermal.rst} (80%)
diff --git a/Documentation/thermal/cpu-cooling-api.txt b/Documentation/thermal/cpu-cooling-api.rst
similarity index 82%
rename from Documentation/thermal/cpu-cooling-api.txt
rename to Documentation/thermal/cpu-cooling-api.rst
index 7df567eaea1a..645d914c45a6 100644
--- a/Documentation/thermal/cpu-cooling-api.txt
+++ b/Documentation/thermal/cpu-cooling-api.rst
@@ -1,5 +1,6 @@
+=======================
CPU cooling APIs How To
-===================================
+=======================
Written by Amit Daniel Kachhap <amit.kachhap@linaro.org>
@@ -8,40 +9,54 @@ Updated: 6 Jan 2015
Copyright (c) 2012 Samsung Electronics Co., Ltd(http://www.samsung.com)
0. Introduction
+===============
The generic cpu cooling(freq clipping) provides registration/unregistration APIs
to the caller. The binding of the cooling devices to the trip point is left for
the user. The registration APIs returns the cooling device pointer.
1. cpu cooling APIs
+===================
1.1 cpufreq registration/unregistration APIs
-1.1.1 struct thermal_cooling_device *cpufreq_cooling_register(
- struct cpumask *clip_cpus)
+--------------------------------------------
+
+ ::
+
+ struct thermal_cooling_device
+ *cpufreq_cooling_register(struct cpumask *clip_cpus)
This interface function registers the cpufreq cooling device with the name
"thermal-cpufreq-%x". This api can support multiple instances of cpufreq
cooling devices.
- clip_cpus: cpumask of cpus where the frequency constraints will happen.
+ clip_cpus:
+ cpumask of cpus where the frequency constraints will happen.
-1.1.2 struct thermal_cooling_device *of_cpufreq_cooling_register(
- struct cpufreq_policy *policy)
+ ::
+
+ struct thermal_cooling_device
+ *of_cpufreq_cooling_register(struct cpufreq_policy *policy)
This interface function registers the cpufreq cooling device with
the name "thermal-cpufreq-%x" linking it with a device tree node, in
order to bind it via the thermal DT code. This api can support multiple
instances of cpufreq cooling devices.
- policy: CPUFreq policy.
+ policy:
+ CPUFreq policy.
-1.1.3 void cpufreq_cooling_unregister(struct thermal_cooling_device *cdev)
+
+ ::
+
+ void cpufreq_cooling_unregister(struct thermal_cooling_device *cdev)
This interface function unregisters the "thermal-cpufreq-%x" cooling device.
cdev: Cooling device pointer which has to be unregistered.
2. Power models
+===============
The power API registration functions provide a simple power model for
CPUs. The current power is calculated as dynamic power (static power isn't
@@ -65,9 +80,9 @@ For a given processor implementation the primary factors are:
variation. In pathological cases this variation can be significant,
but typically it is of a much lesser impact than the factors above.
-A high level dynamic power consumption model may then be represented as:
+A high level dynamic power consumption model may then be represented as::
-Pdyn = f(run) * Voltage^2 * Frequency * Utilisation
+ Pdyn = f(run) * Voltage^2 * Frequency * Utilisation
f(run) here represents the described execution behaviour and its
result has a units of Watts/Hz/Volt^2 (this often expressed in
@@ -80,9 +95,9 @@ factors. Therefore, in initial implementation that contribution is
represented as a constant coefficient. This is a simplification
consistent with the relative contribution to overall power variation.
-In this simplified representation our model becomes:
+In this simplified representation our model becomes::
-Pdyn = Capacitance * Voltage^2 * Frequency * Utilisation
+ Pdyn = Capacitance * Voltage^2 * Frequency * Utilisation
Where `capacitance` is a constant that represents an indicative
running time dynamic power coefficient in fundamental units of
diff --git a/Documentation/thermal/exynos_thermal b/Documentation/thermal/exynos_thermal.rst
similarity index 67%
rename from Documentation/thermal/exynos_thermal
rename to Documentation/thermal/exynos_thermal.rst
index 9010c4416967..5bd556566c70 100644
--- a/Documentation/thermal/exynos_thermal
+++ b/Documentation/thermal/exynos_thermal.rst
@@ -1,8 +1,11 @@
+========================
Kernel driver exynos_tmu
-=================
+========================
Supported chips:
+
* ARM SAMSUNG EXYNOS4, EXYNOS5 series of SoC
+
Datasheet: Not publicly available
Authors: Donggeun Kim <dg77.kim@samsung.com>
@@ -19,32 +22,39 @@ Temperature can be taken from the temperature code.
There are three equations converting from temperature to temperature code.
The three equations are:
- 1. Two point trimming
+ 1. Two point trimming::
+
Tc = (T - 25) * (TI2 - TI1) / (85 - 25) + TI1
- 2. One point trimming
+ 2. One point trimming::
+
Tc = T + TI1 - 25
- 3. No trimming
+ 3. No trimming::
+
Tc = T + 50
- Tc: Temperature code, T: Temperature,
- TI1: Trimming info for 25 degree Celsius (stored at TRIMINFO register)
+ Tc:
+ Temperature code, T: Temperature,
+ TI1:
+ Trimming info for 25 degree Celsius (stored at TRIMINFO register)
Temperature code measured at 25 degree Celsius which is unchanged
- TI2: Trimming info for 85 degree Celsius (stored at TRIMINFO register)
+ TI2:
+ Trimming info for 85 degree Celsius (stored at TRIMINFO register)
Temperature code measured at 85 degree Celsius which is unchanged
TMU(Thermal Management Unit) in EXYNOS4/5 generates interrupt
when temperature exceeds pre-defined levels.
The maximum number of configurable threshold is five.
-The threshold levels are defined as follows:
+The threshold levels are defined as follows::
+
Level_0: current temperature > trigger_level_0 + threshold
Level_1: current temperature > trigger_level_1 + threshold
Level_2: current temperature > trigger_level_2 + threshold
Level_3: current temperature > trigger_level_3 + threshold
- The threshold and each trigger_level are set
- through the corresponding registers.
+The threshold and each trigger_level are set
+through the corresponding registers.
When an interrupt occurs, this driver notify kernel thermal framework
with the function exynos_report_trigger.
@@ -54,24 +64,27 @@ it can be used to synchronize the cooling action.
TMU driver description:
-----------------------
-The exynos thermal driver is structured as,
+The exynos thermal driver is structured as::
Kernel Core thermal framework
(thermal_core.c, step_wise.c, cpu_cooling.c)
^
|
|
-TMU configuration data -------> TMU Driver <------> Exynos Core thermal wrapper
-(exynos_tmu_data.c) (exynos_tmu.c) (exynos_thermal_common.c)
-(exynos_tmu_data.h) (exynos_tmu.h) (exynos_thermal_common.h)
+ TMU configuration data -----> TMU Driver <----> Exynos Core thermal wrapper
+ (exynos_tmu_data.c) (exynos_tmu.c) (exynos_thermal_common.c)
+ (exynos_tmu_data.h) (exynos_tmu.h) (exynos_thermal_common.h)
-a) TMU configuration data: This consist of TMU register offsets/bitfields
+a) TMU configuration data:
+ This consist of TMU register offsets/bitfields
described through structure exynos_tmu_registers. Also several
other platform data (struct exynos_tmu_platform_data) members
are used to configure the TMU.
-b) TMU driver: This component initialises the TMU controller and sets different
+b) TMU driver:
+ This component initialises the TMU controller and sets different
thresholds. It invokes core thermal implementation with the call
exynos_report_trigger.
-c) Exynos Core thermal wrapper: This provides 3 wrapper function to use the
+c) Exynos Core thermal wrapper:
+ This provides 3 wrapper function to use the
Kernel core thermal framework. They are exynos_unregister_thermal,
exynos_register_thermal and exynos_report_trigger.
diff --git a/Documentation/thermal/exynos_thermal_emulation b/Documentation/thermal/exynos_thermal_emulation
deleted file mode 100644
index b15efec6ca28..000000000000
--- a/Documentation/thermal/exynos_thermal_emulation
+++ /dev/null
@@ -1,53 +0,0 @@
-EXYNOS EMULATION MODE
-========================
-
-Copyright (C) 2012 Samsung Electronics
-
-Written by Jonghwa Lee <jonghwa3.lee@samsung.com>
-
-Description
------------
-
-Exynos 4x12 (4212, 4412) and 5 series provide emulation mode for thermal management unit.
-Thermal emulation mode supports software debug for TMU's operation. User can set temperature
-manually with software code and TMU will read current temperature from user value not from
-sensor's value.
-
-Enabling CONFIG_THERMAL_EMULATION option will make this support available.
-When it's enabled, sysfs node will be created as
-/sys/devices/virtual/thermal/thermal_zone'zone id'/emul_temp.
-
-The sysfs node, 'emul_node', will contain value 0 for the initial state. When you input any
-temperature you want to update to sysfs node, it automatically enable emulation mode and
-current temperature will be changed into it.
-(Exynos also supports user changeable delay time which would be used to delay of
- changing temperature. However, this node only uses same delay of real sensing time, 938us.)
-
-Exynos emulation mode requires synchronous of value changing and enabling. It means when you
-want to update the any value of delay or next temperature, then you have to enable emulation
-mode at the same time. (Or you have to keep the mode enabling.) If you don't, it fails to
-change the value to updated one and just use last succeessful value repeatedly. That's why
-this node gives users the right to change termerpature only. Just one interface makes it more
-simply to use.
-
-Disabling emulation mode only requires writing value 0 to sysfs node.
-
-
-TEMP 120 |
- |
- 100 |
- |
- 80 |
- | +-----------
- 60 | | |
- | +-------------| |
- 40 | | | |
- | | | |
- 20 | | | +----------
- | | | | |
- 0 |______________|_____________|__________|__________|_________
- A A A A TIME
- |<----->| |<----->| |<----->| |
- | 938us | | | | | |
-emulation : 0 50 | 70 | 20 | 0
-current temp : sensor 50 70 20 sensor
diff --git a/Documentation/thermal/exynos_thermal_emulation.rst b/Documentation/thermal/exynos_thermal_emulation.rst
new file mode 100644
index 000000000000..c21d10838bc5
--- /dev/null
+++ b/Documentation/thermal/exynos_thermal_emulation.rst
@@ -0,0 +1,61 @@
+=====================
+Exynos Emulation Mode
+=====================
+
+Copyright (C) 2012 Samsung Electronics
+
+Written by Jonghwa Lee <jonghwa3.lee@samsung.com>
+
+Description
+-----------
+
+Exynos 4x12 (4212, 4412) and 5 series provide emulation mode for thermal
+management unit. Thermal emulation mode supports software debug for
+TMU's operation. User can set temperature manually with software code
+and TMU will read current temperature from user value not from sensor's
+value.
+
+Enabling CONFIG_THERMAL_EMULATION option will make this support
+available. When it's enabled, sysfs node will be created as
+/sys/devices/virtual/thermal/thermal_zone'zone id'/emul_temp.
+
+The sysfs node, 'emul_node', will contain value 0 for the initial state.
+When you input any temperature you want to update to sysfs node, it
+automatically enable emulation mode and current temperature will be
+changed into it.
+
+(Exynos also supports user changeable delay time which would be used to
+delay of changing temperature. However, this node only uses same delay
+of real sensing time, 938us.)
+
+Exynos emulation mode requires synchronous of value changing and
+enabling. It means when you want to update the any value of delay or
+next temperature, then you have to enable emulation mode at the same
+time. (Or you have to keep the mode enabling.) If you don't, it fails to
+change the value to updated one and just use last succeessful value
+repeatedly. That's why this node gives users the right to change
+termerpature only. Just one interface makes it more simply to use.
+
+Disabling emulation mode only requires writing value 0 to sysfs node.
+
+::
+
+
+ TEMP 120 |
+ |
+ 100 |
+ |
+ 80 |
+ | +-----------
+ 60 | | |
+ | +-------------| |
+ 40 | | | |
+ | | | |
+ 20 | | | +----------
+ | | | | |
+ 0 |______________|_____________|__________|__________|_________
+ A A A A TIME
+ |<----->| |<----->| |<----->| |
+ | 938us | | | | | |
+ emulation : 0 50 | 70 | 20 | 0
+ current temp: sensor 50 70 20 sensor
diff --git a/Documentation/thermal/index.rst b/Documentation/thermal/index.rst
new file mode 100644
index 000000000000..8c1c00146cad
--- /dev/null
+++ b/Documentation/thermal/index.rst
@@ -0,0 +1,18 @@
+:orphan:
+
+=======
+Thermal
+=======
+
+.. toctree::
+ :maxdepth: 1
+
+ cpu-cooling-api
+ sysfs-api
+ power_allocator
+
+ exynos_thermal
+ exynos_thermal_emulation
+ intel_powerclamp
+ nouveau_thermal
+ x86_pkg_temperature_thermal
diff --git a/Documentation/thermal/intel_powerclamp.txt b/Documentation/thermal/intel_powerclamp.rst
similarity index 76%
rename from Documentation/thermal/intel_powerclamp.txt
rename to Documentation/thermal/intel_powerclamp.rst
index b5df21168fbc..3f6dfb0b3ea6 100644
--- a/Documentation/thermal/intel_powerclamp.txt
+++ b/Documentation/thermal/intel_powerclamp.rst
@@ -1,10 +1,13 @@
- =======================
- INTEL POWERCLAMP DRIVER
- =======================
-By: Arjan van de Ven <arjan@linux.intel.com>
- Jacob Pan <jacob.jun.pan@linux.intel.com>
+=======================
+Intel Powerclamp Driver
+=======================
+
+By:
+ - Arjan van de Ven <arjan@linux.intel.com>
+ - Jacob Pan <jacob.jun.pan@linux.intel.com>
+
+.. Contents:
-Contents:
(*) Introduction
- Goals and Objectives
@@ -23,7 +26,6 @@ Contents:
- Generic Thermal Layer (sysfs)
- Kernel APIs (TBD)
-============
INTRODUCTION
============
@@ -47,7 +49,6 @@ scalability, and user experience. In many cases, clear advantage is
shown over taking the CPU offline or modulating the CPU clock.
-===================
THEORY OF OPERATION
===================
@@ -57,11 +58,12 @@ Idle Injection
On modern Intel processors (Nehalem or later), package level C-state
residency is available in MSRs, thus also available to the kernel.
-These MSRs are:
- #define MSR_PKG_C2_RESIDENCY 0x60D
- #define MSR_PKG_C3_RESIDENCY 0x3F8
- #define MSR_PKG_C6_RESIDENCY 0x3F9
- #define MSR_PKG_C7_RESIDENCY 0x3FA
+These MSRs are::
+
+ #define MSR_PKG_C2_RESIDENCY 0x60D
+ #define MSR_PKG_C3_RESIDENCY 0x3F8
+ #define MSR_PKG_C6_RESIDENCY 0x3F9
+ #define MSR_PKG_C7_RESIDENCY 0x3FA
If the kernel can also inject idle time to the system, then a
closed-loop control system can be established that manages package
@@ -96,19 +98,21 @@ are not masked. Tests show that the extra wakeups from scheduler tick
have a dramatic impact on the effectiveness of the powerclamp driver
on large scale systems (Westmere system with 80 processors).
-CPU0
- ____________ ____________
-kidle_inject/0 | sleep | mwait | sleep |
- _________| |________| |_______
- duration
-CPU1
- ____________ ____________
-kidle_inject/1 | sleep | mwait | sleep |
- _________| |________| |_______
- ^
- |
- |
- roundup(jiffies, interval)
+::
+
+ CPU0
+ ____________ ____________
+ kidle_inject/0 | sleep | mwait | sleep |
+ _________| |________| |_______
+ duration
+ CPU1
+ ____________ ____________
+ kidle_inject/1 | sleep | mwait | sleep |
+ _________| |________| |_______
+ ^
+ |
+ |
+ roundup(jiffies, interval)
Only one CPU is allowed to collect statistics and update global
control parameters. This CPU is referred to as the controlling CPU in
@@ -148,7 +152,7 @@ b) determine the amount of compensation needed at each target ratio
Compensation to each target ratio consists of two parts:
- a) steady state error compensation
+ a) steady state error compensation
This is to offset the error occurring when the system can
enter idle without extra wakeups (such as external interrupts).
@@ -158,41 +162,42 @@ Compensation to each target ratio consists of two parts:
slowing down CPU activities.
A debugfs file is provided for the user to examine compensation
-progress and results, such as on a Westmere system.
-[jacob@nex01 ~]$ cat
-/sys/kernel/debug/intel_powerclamp/powerclamp_calib
-controlling cpu: 0
-pct confidence steady dynamic (compensation)
-0 0 0 0
-1 1 0 0
-2 1 1 0
-3 3 1 0
-4 3 1 0
-5 3 1 0
-6 3 1 0
-7 3 1 0
-8 3 1 0
-...
-30 3 2 0
-31 3 2 0
-32 3 1 0
-33 3 2 0
-34 3 1 0
-35 3 2 0
-36 3 1 0
-37 3 2 0
-38 3 1 0
-39 3 2 0
-40 3 3 0
-41 3 1 0
-42 3 2 0
-43 3 1 0
-44 3 1 0
-45 3 2 0
-46 3 3 0
-47 3 0 0
-48 3 2 0
-49 3 3 0
+progress and results, such as on a Westmere system::
+
+ [jacob@nex01 ~]$ cat
+ /sys/kernel/debug/intel_powerclamp/powerclamp_calib
+ controlling cpu: 0
+ pct confidence steady dynamic (compensation)
+ 0 0 0 0
+ 1 1 0 0
+ 2 1 1 0
+ 3 3 1 0
+ 4 3 1 0
+ 5 3 1 0
+ 6 3 1 0
+ 7 3 1 0
+ 8 3 1 0
+ ...
+ 30 3 2 0
+ 31 3 2 0
+ 32 3 1 0
+ 33 3 2 0
+ 34 3 1 0
+ 35 3 2 0
+ 36 3 1 0
+ 37 3 2 0
+ 38 3 1 0
+ 39 3 2 0
+ 40 3 3 0
+ 41 3 1 0
+ 42 3 2 0
+ 43 3 1 0
+ 44 3 1 0
+ 45 3 2 0
+ 46 3 3 0
+ 47 3 0 0
+ 48 3 2 0
+ 49 3 3 0
Calibration occurs during runtime. No offline method is available.
Steady state compensation is used only when confidence levels of all
@@ -217,9 +222,8 @@ keeps track of clamping kernel threads, even after they are migrated
to other CPUs, after a CPU offline event.
-=====================
Performance Analysis
-=====================
+====================
This section describes the general performance data collected on
multiple systems, including Westmere (80P) and Ivy Bridge (4P, 8P).
@@ -257,16 +261,15 @@ achieve up to 40% better performance per watt. (measured by a spin
counter summed over per CPU counting threads spawned for all running
CPUs).
-====================
Usage and Interfaces
====================
The powerclamp driver is registered to the generic thermal layer as a
-cooling device. Currently, it’s not bound to any thermal zones.
+cooling device. Currently, it’s not bound to any thermal zones::
-jacob@chromoly:/sys/class/thermal/cooling_device14$ grep . *
-cur_state:0
-max_state:50
-type:intel_powerclamp
+ jacob@chromoly:/sys/class/thermal/cooling_device14$ grep . *
+ cur_state:0
+ max_state:50
+ type:intel_powerclamp
cur_state allows user to set the desired idle percentage. Writing 0 to
cur_state will stop idle injection. Writing a value between 1 and
@@ -278,9 +281,9 @@ cur_state returns value -1 instead of 0 which is to avoid confusing
100% busy state with the disabled state.
Example usage:
-- To inject 25% idle time
-$ sudo sh -c "echo 25 > /sys/class/thermal/cooling_device80/cur_state
-"
+- To inject 25% idle time::
+
+ $ sudo sh -c "echo 25 > /sys/class/thermal/cooling_device80/cur_state
If the system is not busy and has more than 25% idle time already,
then the powerclamp driver will not start idle injection. Using Top
@@ -292,23 +295,23 @@ idle time is accounted as normal idle in that common code path is
taken as the idle task.
In this example, 24.1% idle is shown. This helps the system admin or
-user determine the cause of slowdown, when a powerclamp driver is in action.
+user determine the cause of slowdown, when a powerclamp driver is in action::
-Tasks: 197 total, 1 running, 196 sleeping, 0 stopped, 0 zombie
-Cpu(s): 71.2%us, 4.7%sy, 0.0%ni, 24.1%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
-Mem: 3943228k total, 1689632k used, 2253596k free, 74960k buffers
-Swap: 4087804k total, 0k used, 4087804k free, 945336k cached
+ Tasks: 197 total, 1 running, 196 sleeping, 0 stopped, 0 zombie
+ Cpu(s): 71.2%us, 4.7%sy, 0.0%ni, 24.1%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
+ Mem: 3943228k total, 1689632k used, 2253596k free, 74960k buffers
+ Swap: 4087804k total, 0k used, 4087804k free, 945336k cached
- PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
- 3352 jacob 20 0 262m 644 428 S 286 0.0 0:17.16 spin
- 3341 root -51 0 0 0 0 D 25 0.0 0:01.62 kidle_inject/0
- 3344 root -51 0 0 0 0 D 25 0.0 0:01.60 kidle_inject/3
- 3342 root -51 0 0 0 0 D 25 0.0 0:01.61 kidle_inject/1
- 3343 root -51 0 0 0 0 D 25 0.0 0:01.60 kidle_inject/2
- 2935 jacob 20 0 696m 125m 35m S 5 3.3 0:31.11 firefox
- 1546 root 20 0 158m 20m 6640 S 3 0.5 0:26.97 Xorg
- 2100 jacob 20 0 1223m 88m 30m S 3 2.3 0:23.68 compiz
+ PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
+ 3352 jacob 20 0 262m 644 428 S 286 0.0 0:17.16 spin
+ 3341 root -51 0 0 0 0 D 25 0.0 0:01.62 kidle_inject/0
+ 3344 root -51 0 0 0 0 D 25 0.0 0:01.60 kidle_inject/3
+ 3342 root -51 0 0 0 0 D 25 0.0 0:01.61 kidle_inject/1
+ 3343 root -51 0 0 0 0 D 25 0.0 0:01.60 kidle_inject/2
+ 2935 jacob 20 0 696m 125m 35m S 5 3.3 0:31.11 firefox
+ 1546 root 20 0 158m 20m 6640 S 3 0.5 0:26.97 Xorg
+ 2100 jacob 20 0 1223m 88m 30m S 3 2.3 0:23.68 compiz
Tests have shown that by using the powerclamp driver as a cooling
device, a PID based userspace thermal controller can manage to
diff --git a/Documentation/thermal/nouveau_thermal b/Documentation/thermal/nouveau_thermal.rst
similarity index 64%
rename from Documentation/thermal/nouveau_thermal
rename to Documentation/thermal/nouveau_thermal.rst
index 6e17a11efcb0..37255fd6735d 100644
--- a/Documentation/thermal/nouveau_thermal
+++ b/Documentation/thermal/nouveau_thermal.rst
@@ -1,13 +1,15 @@
+=====================
Kernel driver nouveau
-===================
+=====================
Supported chips:
+
* NV43+
Authors: Martin Peres (mupuf) <martin.peres@free.fr>
Description
----------
+-----------
This driver allows to read the GPU core temperature, drive the GPU fan and
set temperature alarms.
@@ -19,20 +21,25 @@ interface is likely not to work. This document may then not cover your situation
entirely.
Temperature management
---------------------
+----------------------
Temperature is exposed under as a read-only HWMON attribute temp1_input.
In order to protect the GPU from overheating, Nouveau supports 4 configurable
temperature thresholds:
- * Fan_boost: Fan speed is set to 100% when reaching this temperature;
- * Downclock: The GPU will be downclocked to reduce its power dissipation;
- * Critical: The GPU is put on hold to further lower power dissipation;
- * Shutdown: Shut the computer down to protect your GPU.
+ * Fan_boost:
+ Fan speed is set to 100% when reaching this temperature;
+ * Downclock:
+ The GPU will be downclocked to reduce its power dissipation;
+ * Critical:
+ The GPU is put on hold to further lower power dissipation;
+ * Shutdown:
+ Shut the computer down to protect your GPU.
-WARNING: Some of these thresholds may not be used by Nouveau depending
-on your chipset.
+WARNING:
+ Some of these thresholds may not be used by Nouveau depending
+ on your chipset.
The default value for these thresholds comes from the GPU's vbios. These
thresholds can be configured thanks to the following HWMON attributes:
@@ -46,19 +53,24 @@ NOTE: Remember that the values are stored as milli degrees Celsius. Don't forget
to multiply!
Fan management
-------------
+--------------
Not all cards have a drivable fan. If you do, then the following HWMON
attributes should be available:
- * pwm1_enable: Current fan management mode (NONE, MANUAL or AUTO);
- * pwm1: Current PWM value (power percentage);
- * pwm1_min: The minimum PWM speed allowed;
- * pwm1_max: The maximum PWM speed allowed (bypassed when hitting Fan_boost);
+ * pwm1_enable:
+ Current fan management mode (NONE, MANUAL or AUTO);
+ * pwm1:
+ Current PWM value (power percentage);
+ * pwm1_min:
+ The minimum PWM speed allowed;
+ * pwm1_max:
+ The maximum PWM speed allowed (bypassed when hitting Fan_boost);
You may also have the following attribute:
- * fan1_input: Speed in RPM of your fan.
+ * fan1_input:
+ Speed in RPM of your fan.
Your fan can be driven in different modes:
@@ -66,14 +78,16 @@ Your fan can be driven in different modes:
* 1: The fan can be driven in manual (use pwm1 to change the speed);
* 2; The fan is driven automatically depending on the temperature.
-NOTE: Be sure to use the manual mode if you want to drive the fan speed manually
+NOTE:
+ Be sure to use the manual mode if you want to drive the fan speed manually
-NOTE2: When operating in manual mode outside the vbios-defined
-[PWM_min, PWM_max] range, the reported fan speed (RPM) may not be accurate
-depending on your hardware.
+NOTE2:
+ When operating in manual mode outside the vbios-defined
+ [PWM_min, PWM_max] range, the reported fan speed (RPM) may not be accurate
+ depending on your hardware.
Bug reports
----------
+-----------
Thermal management on Nouveau is new and may not work on all cards. If you have
inquiries, please ping mupuf on IRC (#nouveau, freenode).
diff --git a/Documentation/thermal/power_allocator.txt b/Documentation/thermal/power_allocator.rst
similarity index 74%
rename from Documentation/thermal/power_allocator.txt
rename to Documentation/thermal/power_allocator.rst
index 9fb0ff06dca9..67b6a3297238 100644
--- a/Documentation/thermal/power_allocator.txt
+++ b/Documentation/thermal/power_allocator.rst
@@ -1,3 +1,4 @@
+=================================
Power allocator governor tunables
=================================
@@ -25,36 +26,36 @@ temperature as the control input and power as the controlled output:
P_max = k_p * e + k_i * err_integral + k_d * diff_err + sustainable_power
where
- e = desired_temperature - current_temperature
- err_integral is the sum of previous errors
- diff_err = e - previous_error
+ - e = desired_temperature - current_temperature
+ - err_integral is the sum of previous errors
+ - diff_err = e - previous_error
-It is similar to the one depicted below:
+It is similar to the one depicted below::
- k_d
- |
-current_temp |
- | v
- | +----------+ +---+
- | +----->| diff_err |-->| X |------+
- | | +----------+ +---+ |
- | | | tdp actor
- | | k_i | | get_requested_power()
- | | | | | | |
- | | | | | | | ...
- v | v v v v v
- +---+ | +-------+ +---+ +---+ +---+ +----------+
- | S |-------+----->| sum e |----->| X |--->| S |-->| S |-->|power |
- +---+ | +-------+ +---+ +---+ +---+ |allocation|
- ^ | ^ +----------+
- | | | | |
- | | +---+ | | |
- | +------->| X |-------------------+ v v
- | +---+ granted performance
-desired_temperature ^
- |
- |
- k_po/k_pu
+ k_d
+ |
+ current_temp |
+ | v
+ | +----------+ +---+
+ | +----->| diff_err |-->| X |------+
+ | | +----------+ +---+ |
+ | | | tdp actor
+ | | k_i | | get_requested_power()
+ | | | | | | |
+ | | | | | | | ...
+ v | v v v v v
+ +---+ | +-------+ +---+ +---+ +---+ +----------+
+ | S |-----+----->| sum e |----->| X |--->| S |-->| S |-->|power |
+ +---+ | +-------+ +---+ +---+ +---+ |allocation|
+ ^ | ^ +----------+
+ | | | | |
+ | | +---+ | | |
+ | +------->| X |-------------------+ v v
+ | +---+ granted performance
+ desired_temperature ^
+ |
+ |
+ k_po/k_pu
Sustainable power
-----------------
@@ -73,7 +74,7 @@ is typically 2000mW, while on a 10" tablet is around 4500mW (may vary
depending on screen size).
If you are using device tree, do add it as a property of the
-thermal-zone. For example:
+thermal-zone. For example::
thermal-zones {
soc_thermal {
@@ -85,7 +86,7 @@ thermal-zone. For example:
Instead, if the thermal zone is registered from the platform code, pass a
`thermal_zone_params` that has a `sustainable_power`. If no
`thermal_zone_params` were being passed, then something like below
-will suffice:
+will suffice::
static const struct thermal_zone_params tz_params = {
.sustainable_power = 3500,
@@ -112,18 +113,18 @@ available capacity at a low temperature. On the other hand, a high
value of `k_pu` will result in the governor granting very high power
while temperature is low, and may lead to temperature overshooting.
-The default value for `k_pu` is:
+The default value for `k_pu` is::
2 * sustainable_power / (desired_temperature - switch_on_temp)
This means that at `switch_on_temp` the output of the controller's
proportional term will be 2 * `sustainable_power`. The default value
-for `k_po` is:
+for `k_po` is::
sustainable_power / (desired_temperature - switch_on_temp)
Focusing on the proportional and feed forward values of the PID
-controller equation we have:
+controller equation we have::
P_max = k_p * e + sustainable_power
@@ -134,21 +135,23 @@ is the desired one, then the proportional component is zero and
thermal equilibrium under constant load. `sustainable_power` is only
an estimate, which is the reason for closed-loop control such as this.
-Expanding `k_pu` we get:
+Expanding `k_pu` we get::
+
P_max = 2 * sustainable_power * (T_set - T) / (T_set - T_on) +
- sustainable_power
+ sustainable_power
-where
- T_set is the desired temperature
- T is the current temperature
- T_on is the switch on temperature
+where:
+
+ - T_set is the desired temperature
+ - T is the current temperature
+ - T_on is the switch on temperature
When the current temperature is the switch_on temperature, the above
-formula becomes:
+formula becomes::
P_max = 2 * sustainable_power * (T_set - T_on) / (T_set - T_on) +
- sustainable_power = 2 * sustainable_power + sustainable_power =
- 3 * sustainable_power
+ sustainable_power = 2 * sustainable_power + sustainable_power =
+ 3 * sustainable_power
Therefore, the proportional term alone linearly decreases power from
3 * `sustainable_power` to `sustainable_power` as the temperature
@@ -178,11 +181,18 @@ Cooling device power API
Cooling devices controlled by this governor must supply the additional
"power" API in their `cooling_device_ops`. It consists on three ops:
-1. int get_requested_power(struct thermal_cooling_device *cdev,
- struct thermal_zone_device *tz, u32 *power);
-@cdev: The `struct thermal_cooling_device` pointer
-@tz: thermal zone in which we are currently operating
-@power: pointer in which to store the calculated power
+1. ::
+
+ int get_requested_power(struct thermal_cooling_device *cdev,
+ struct thermal_zone_device *tz, u32 *power);
+
+
+@cdev:
+ The `struct thermal_cooling_device` pointer
+@tz:
+ thermal zone in which we are currently operating
+@power:
+ pointer in which to store the calculated power
`get_requested_power()` calculates the power requested by the device
in milliwatts and stores it in @power . It should return 0 on
@@ -190,23 +200,37 @@ success, -E* on failure. This is currently used by the power
allocator governor to calculate how much power to give to each cooling
device.
-2. int state2power(struct thermal_cooling_device *cdev, struct
- thermal_zone_device *tz, unsigned long state, u32 *power);
-@cdev: The `struct thermal_cooling_device` pointer
-@tz: thermal zone in which we are currently operating
-@state: A cooling device state
-@power: pointer in which to store the equivalent power
+2. ::
+
+ int state2power(struct thermal_cooling_device *cdev, struct
+ thermal_zone_device *tz, unsigned long state,
+ u32 *power);
+
+@cdev:
+ The `struct thermal_cooling_device` pointer
+@tz:
+ thermal zone in which we are currently operating
+@state:
+ A cooling device state
+@power:
+ pointer in which to store the equivalent power
Convert cooling device state @state into power consumption in
milliwatts and store it in @power. It should return 0 on success, -E*
on failure. This is currently used by thermal core to calculate the
maximum power that an actor can consume.
-3. int power2state(struct thermal_cooling_device *cdev, u32 power,
- unsigned long *state);
-@cdev: The `struct thermal_cooling_device` pointer
-@power: power in milliwatts
-@state: pointer in which to store the resulting state
+3. ::
+
+ int power2state(struct thermal_cooling_device *cdev, u32 power,
+ unsigned long *state);
+
+@cdev:
+ The `struct thermal_cooling_device` pointer
+@power:
+ power in milliwatts
+@state:
+ pointer in which to store the resulting state
Calculate a cooling device state that would make the device consume at
most @power mW and store it in @state. It should return 0 on success,
diff --git a/Documentation/thermal/sysfs-api.txt b/Documentation/thermal/sysfs-api.rst
similarity index 66%
rename from Documentation/thermal/sysfs-api.txt
rename to Documentation/thermal/sysfs-api.rst
index c3fa500df92c..e4930761d3e5 100644
--- a/Documentation/thermal/sysfs-api.txt
+++ b/Documentation/thermal/sysfs-api.rst
@@ -1,3 +1,4 @@
+===================================
Generic Thermal Sysfs driver How To
===================================
@@ -9,6 +10,7 @@ Copyright (c) 2008 Intel Corporation
0. Introduction
+===============
The generic thermal sysfs provides a set of interfaces for thermal zone
devices (sensors) and thermal cooling devices (fan, processor...) to register
@@ -25,59 +27,90 @@ An intelligent thermal management application can make decisions based on
inputs from thermal zone attributes (the current temperature and trip point
temperature) and throttle appropriate devices.
-[0-*] denotes any positive number starting from 0
-[1-*] denotes any positive number starting from 1
+- `[0-*]` denotes any positive number starting from 0
+- `[1-*]` denotes any positive number starting from 1
1. thermal sysfs driver interface functions
+===========================================
1.1 thermal zone device interface
-1.1.1 struct thermal_zone_device *thermal_zone_device_register(char *type,
- int trips, int mask, void *devdata,
- struct thermal_zone_device_ops *ops,
- const struct thermal_zone_params *tzp,
- int passive_delay, int polling_delay))
+---------------------------------
+
+ ::
+
+ struct thermal_zone_device
+ *thermal_zone_device_register(char *type,
+ int trips, int mask, void *devdata,
+ struct thermal_zone_device_ops *ops,
+ const struct thermal_zone_params *tzp,
+ int passive_delay, int polling_delay))
This interface function adds a new thermal zone device (sensor) to
- /sys/class/thermal folder as thermal_zone[0-*]. It tries to bind all the
+ /sys/class/thermal folder as `thermal_zone[0-*]`. It tries to bind all the
thermal cooling devices registered at the same time.
- type: the thermal zone type.
- trips: the total number of trip points this thermal zone supports.
- mask: Bit string: If 'n'th bit is set, then trip point 'n' is writeable.
- devdata: device private data
- ops: thermal zone device call-backs.
- .bind: bind the thermal zone device with a thermal cooling device.
- .unbind: unbind the thermal zone device with a thermal cooling device.
- .get_temp: get the current temperature of the thermal zone.
- .set_trips: set the trip points window. Whenever the current temperature
+ type:
+ the thermal zone type.
+ trips:
+ the total number of trip points this thermal zone supports.
+ mask:
+ Bit string: If 'n'th bit is set, then trip point 'n' is writeable.
+ devdata:
+ device private data
+ ops:
+ thermal zone device call-backs.
+
+ .bind:
+ bind the thermal zone device with a thermal cooling device.
+ .unbind:
+ unbind the thermal zone device with a thermal cooling device.
+ .get_temp:
+ get the current temperature of the thermal zone.
+ .set_trips:
+ set the trip points window. Whenever the current temperature
is updated, the trip points immediately below and above the
current temperature are found.
- .get_mode: get the current mode (enabled/disabled) of the thermal zone.
- - "enabled" means the kernel thermal management is enabled.
- - "disabled" will prevent kernel thermal driver action upon trip points
- so that user applications can take charge of thermal management.
- .set_mode: set the mode (enabled/disabled) of the thermal zone.
- .get_trip_type: get the type of certain trip point.
- .get_trip_temp: get the temperature above which the certain trip point
+ .get_mode:
+ get the current mode (enabled/disabled) of the thermal zone.
+
+ - "enabled" means the kernel thermal management is
+ enabled.
+ - "disabled" will prevent kernel thermal driver action
+ upon trip points so that user applications can take
+ charge of thermal management.
+ .set_mode:
+ set the mode (enabled/disabled) of the thermal zone.
+ .get_trip_type:
+ get the type of certain trip point.
+ .get_trip_temp:
+ get the temperature above which the certain trip point
will be fired.
- .set_emul_temp: set the emulation temperature which helps in debugging
+ .set_emul_temp:
+ set the emulation temperature which helps in debugging
different threshold temperature points.
- tzp: thermal zone platform parameters.
- passive_delay: number of milliseconds to wait between polls when
+ tzp:
+ thermal zone platform parameters.
+ passive_delay:
+ number of milliseconds to wait between polls when
performing passive cooling.
- polling_delay: number of milliseconds to wait between polls when checking
+ polling_delay:
+ number of milliseconds to wait between polls when checking
whether trip points have been crossed (0 for interrupt driven systems).
+ ::
-1.1.2 void thermal_zone_device_unregister(struct thermal_zone_device *tz)
+ void thermal_zone_device_unregister(struct thermal_zone_device *tz)
This interface function removes the thermal zone device.
It deletes the corresponding entry from /sys/class/thermal folder and
unbinds all the thermal cooling devices it uses.
-1.1.3 struct thermal_zone_device *thermal_zone_of_sensor_register(
- struct device *dev, int sensor_id, void *data,
- const struct thermal_zone_of_device_ops *ops)
+ ::
+
+ struct thermal_zone_device
+ *thermal_zone_of_sensor_register(struct device *dev, int sensor_id,
+ void *data,
+ const struct thermal_zone_of_device_ops *ops)
This interface adds a new sensor to a DT thermal zone.
This function will search the list of thermal zones described in
@@ -87,25 +120,33 @@ temperature) and throttle appropriate devices.
thermal zone device.
The parameters for this interface are:
- dev: Device node of sensor containing valid node pointer in
+
+ dev:
+ Device node of sensor containing valid node pointer in
dev->of_node.
- sensor_id: a sensor identifier, in case the sensor IP has more
+ sensor_id:
+ a sensor identifier, in case the sensor IP has more
than one sensors
- data: a private pointer (owned by the caller) that will be
+ data:
+ a private pointer (owned by the caller) that will be
passed back, when a temperature reading is needed.
- ops: struct thermal_zone_of_device_ops *.
+ ops:
+ `struct thermal_zone_of_device_ops *`.
- get_temp: a pointer to a function that reads the
+ ============== =======================================
+ get_temp a pointer to a function that reads the
sensor temperature. This is mandatory
callback provided by sensor driver.
- set_trips: a pointer to a function that sets a
+ set_trips a pointer to a function that sets a
temperature window. When this window is
left the driver must inform the thermal
core via thermal_zone_device_update.
- get_trend: a pointer to a function that reads the
+ get_trend a pointer to a function that reads the
sensor temperature trend.
- set_emul_temp: a pointer to a function that sets
+ set_emul_temp a pointer to a function that sets
sensor emulated temperature.
+ ============== =======================================
+
The thermal zone temperature is provided by the get_temp() function
pointer of thermal_zone_of_device_ops. When called, it will
have the private pointer @data back.
@@ -114,8 +155,10 @@ temperature) and throttle appropriate devices.
handle. Caller should check the return handle with IS_ERR() for finding
whether success or not.
-1.1.4 void thermal_zone_of_sensor_unregister(struct device *dev,
- struct thermal_zone_device *tzd)
+ ::
+
+ void thermal_zone_of_sensor_unregister(struct device *dev,
+ struct thermal_zone_device *tzd)
This interface unregisters a sensor from a DT thermal zone which was
successfully added by interface thermal_zone_of_sensor_register().
@@ -124,21 +167,29 @@ temperature) and throttle appropriate devices.
interface. It will also silent the zone by remove the .get_temp() and
get_trend() thermal zone device callbacks.
-1.1.5 struct thermal_zone_device *devm_thermal_zone_of_sensor_register(
- struct device *dev, int sensor_id,
- void *data, const struct thermal_zone_of_device_ops *ops)
+ ::
+
+ struct thermal_zone_device
+ *devm_thermal_zone_of_sensor_register(struct device *dev,
+ int sensor_id,
+ void *data,
+ const struct thermal_zone_of_device_ops *ops)
This interface is resource managed version of
thermal_zone_of_sensor_register().
+
All details of thermal_zone_of_sensor_register() described in
section 1.1.3 is applicable here.
+
The benefit of using this interface to register sensor is that it
is not require to explicitly call thermal_zone_of_sensor_unregister()
in error path or during driver unbinding as this is done by driver
resource manager.
-1.1.6 void devm_thermal_zone_of_sensor_unregister(struct device *dev,
- struct thermal_zone_device *tzd)
+ ::
+
+ void devm_thermal_zone_of_sensor_unregister(struct device *dev,
+ struct thermal_zone_device *tzd)
This interface is resource managed version of
thermal_zone_of_sensor_unregister().
@@ -147,123 +198,186 @@ temperature) and throttle appropriate devices.
Normally this function will not need to be called and the resource
management code will ensure that the resource is freed.
-1.1.7 int thermal_zone_get_slope(struct thermal_zone_device *tz)
+ ::
+
+ int thermal_zone_get_slope(struct thermal_zone_device *tz)
This interface is used to read the slope attribute value
for the thermal zone device, which might be useful for platform
drivers for temperature calculations.
-1.1.8 int thermal_zone_get_offset(struct thermal_zone_device *tz)
+ ::
+
+ int thermal_zone_get_offset(struct thermal_zone_device *tz)
This interface is used to read the offset attribute value
for the thermal zone device, which might be useful for platform
drivers for temperature calculations.
1.2 thermal cooling device interface
-1.2.1 struct thermal_cooling_device *thermal_cooling_device_register(char *name,
- void *devdata, struct thermal_cooling_device_ops *)
+------------------------------------
+
+
+ ::
+
+ struct thermal_cooling_device
+ *thermal_cooling_device_register(char *name,
+ void *devdata, struct thermal_cooling_device_ops *)
This interface function adds a new thermal cooling device (fan/processor/...)
- to /sys/class/thermal/ folder as cooling_device[0-*]. It tries to bind itself
+ to /sys/class/thermal/ folder as `cooling_device[0-*]`. It tries to bind itself
to all the thermal zone devices registered at the same time.
- name: the cooling device name.
- devdata: device private data.
- ops: thermal cooling devices call-backs.
- .get_max_state: get the Maximum throttle state of the cooling device.
- .get_cur_state: get the Currently requested throttle state of the cooling device.
- .set_cur_state: set the Current throttle state of the cooling device.
-
-1.2.2 void thermal_cooling_device_unregister(struct thermal_cooling_device *cdev)
+
+ name:
+ the cooling device name.
+ devdata:
+ device private data.
+ ops:
+ thermal cooling devices call-backs.
+
+ .get_max_state:
+ get the Maximum throttle state of the cooling device.
+ .get_cur_state:
+ get the Currently requested throttle state of the
+ cooling device.
+ .set_cur_state:
+ set the Current throttle state of the cooling device.
+
+ ::
+
+ void thermal_cooling_device_unregister(struct thermal_cooling_device *cdev)
This interface function removes the thermal cooling device.
It deletes the corresponding entry from /sys/class/thermal folder and
unbinds itself from all the thermal zone devices using it.
1.3 interface for binding a thermal zone device with a thermal cooling device
-1.3.1 int thermal_zone_bind_cooling_device(struct thermal_zone_device *tz,
- int trip, struct thermal_cooling_device *cdev,
- unsigned long upper, unsigned long lower, unsigned int weight);
+-----------------------------------------------------------------------------
+
+ ::
+
+ int thermal_zone_bind_cooling_device(struct thermal_zone_device *tz,
+ int trip, struct thermal_cooling_device *cdev,
+ unsigned long upper, unsigned long lower, unsigned int weight);
This interface function binds a thermal cooling device to a particular trip
point of a thermal zone device.
+
This function is usually called in the thermal zone device .bind callback.
- tz: the thermal zone device
- cdev: thermal cooling device
- trip: indicates which trip point in this thermal zone the cooling device
- is associated with.
- upper:the Maximum cooling state for this trip point.
- THERMAL_NO_LIMIT means no upper limit,
+
+ tz:
+ the thermal zone device
+ cdev:
+ thermal cooling device
+ trip:
+ indicates which trip point in this thermal zone the cooling device
+ is associated with.
+ upper:
+ the Maximum cooling state for this trip point.
+ THERMAL_NO_LIMIT means no upper limit,
and the cooling device can be in max_state.
- lower:the Minimum cooling state can be used for this trip point.
- THERMAL_NO_LIMIT means no lower limit,
+ lower:
+ the Minimum cooling state can be used for this trip point.
+ THERMAL_NO_LIMIT means no lower limit,
and the cooling device can be in cooling state 0.
- weight: the influence of this cooling device in this thermal
- zone. See 1.4.1 below for more information.
+ weight:
+ the influence of this cooling device in this thermal
+ zone. See 1.4.1 below for more information.
-1.3.2 int thermal_zone_unbind_cooling_device(struct thermal_zone_device *tz,
- int trip, struct thermal_cooling_device *cdev);
+ ::
+
+ int thermal_zone_unbind_cooling_device(struct thermal_zone_device *tz,
+ int trip, struct thermal_cooling_device *cdev);
This interface function unbinds a thermal cooling device from a particular
trip point of a thermal zone device. This function is usually called in
the thermal zone device .unbind callback.
- tz: the thermal zone device
- cdev: thermal cooling device
- trip: indicates which trip point in this thermal zone the cooling device
- is associated with.
+
+ tz:
+ the thermal zone device
+ cdev:
+ thermal cooling device
+ trip:
+ indicates which trip point in this thermal zone the cooling device
+ is associated with.
1.4 Thermal Zone Parameters
-1.4.1 struct thermal_bind_params
+---------------------------
+
+ ::
+
+ struct thermal_bind_params
+
This structure defines the following parameters that are used to bind
a zone with a cooling device for a particular trip point.
- .cdev: The cooling device pointer
- .weight: The 'influence' of a particular cooling device on this
- zone. This is relative to the rest of the cooling
- devices. For example, if all cooling devices have a
- weight of 1, then they all contribute the same. You can
- use percentages if you want, but it's not mandatory. A
- weight of 0 means that this cooling device doesn't
- contribute to the cooling of this zone unless all cooling
- devices have a weight of 0. If all weights are 0, then
- they all contribute the same.
- .trip_mask:This is a bit mask that gives the binding relation between
- this thermal zone and cdev, for a particular trip point.
- If nth bit is set, then the cdev and thermal zone are bound
- for trip point n.
- .binding_limits: This is an array of cooling state limits. Must have
- exactly 2 * thermal_zone.number_of_trip_points. It is an
- array consisting of tuples <lower-state upper-state> of
- state limits. Each trip will be associated with one state
- limit tuple when binding. A NULL pointer means
- <THERMAL_NO_LIMITS THERMAL_NO_LIMITS> on all trips.
- These limits are used when binding a cdev to a trip point.
- .match: This call back returns success(0) if the 'tz and cdev' need to
+
+ .cdev:
+ The cooling device pointer
+ .weight:
+ The 'influence' of a particular cooling device on this
+ zone. This is relative to the rest of the cooling
+ devices. For example, if all cooling devices have a
+ weight of 1, then they all contribute the same. You can
+ use percentages if you want, but it's not mandatory. A
+ weight of 0 means that this cooling device doesn't
+ contribute to the cooling of this zone unless all cooling
+ devices have a weight of 0. If all weights are 0, then
+ they all contribute the same.
+ .trip_mask:
+ This is a bit mask that gives the binding relation between
+ this thermal zone and cdev, for a particular trip point.
+ If nth bit is set, then the cdev and thermal zone are bound
+ for trip point n.
+ .binding_limits:
+ This is an array of cooling state limits. Must have
+ exactly 2 * thermal_zone.number_of_trip_points. It is an
+ array consisting of tuples <lower-state upper-state> of
+ state limits. Each trip will be associated with one state
+ limit tuple when binding. A NULL pointer means
+ <THERMAL_NO_LIMITS THERMAL_NO_LIMITS> on all trips.
+ These limits are used when binding a cdev to a trip point.
+ .match:
+ This call back returns success(0) if the 'tz and cdev' need to
be bound, as per platform data.
-1.4.2 struct thermal_zone_params
+
+ ::
+
+ struct thermal_zone_params
+
This structure defines the platform level parameters for a thermal zone.
This data, for each thermal zone should come from the platform layer.
This is an optional feature where some platforms can choose not to
provide this data.
- .governor_name: Name of the thermal governor used for this zone
- .no_hwmon: a boolean to indicate if the thermal to hwmon sysfs interface
- is required. when no_hwmon == false, a hwmon sysfs interface
- will be created. when no_hwmon == true, nothing will be done.
- In case the thermal_zone_params is NULL, the hwmon interface
- will be created (for backward compatibility).
- .num_tbps: Number of thermal_bind_params entries for this zone
- .tbp: thermal_bind_params entries
+
+ .governor_name:
+ Name of the thermal governor used for this zone
+ .no_hwmon:
+ a boolean to indicate if the thermal to hwmon sysfs interface
+ is required. when no_hwmon == false, a hwmon sysfs interface
+ will be created. when no_hwmon == true, nothing will be done.
+ In case the thermal_zone_params is NULL, the hwmon interface
+ will be created (for backward compatibility).
+ .num_tbps:
+ Number of thermal_bind_params entries for this zone
+ .tbp:
+ thermal_bind_params entries
2. sysfs attributes structure
+=============================
+== ================
RO read only value
WO write only value
RW read/write value
+== ================
Thermal sysfs attributes will be represented under /sys/class/thermal.
Hwmon sysfs I/F extension is also available under /sys/class/hwmon
if hwmon is compiled in or built as a module.
-Thermal zone device sys I/F, created once it's registered:
-/sys/class/thermal/thermal_zone[0-*]:
+Thermal zone device sys I/F, created once it's registered::
+
+ /sys/class/thermal/thermal_zone[0-*]:
|---type: Type of the thermal zone
|---temp: Current temperature
|---mode: Working mode of the thermal zone
@@ -282,8 +396,9 @@ Thermal zone device sys I/F, created once it's registered:
|---slope: Slope constant applied as linear extrapolation
|---offset: Offset constant applied as linear extrapolation
-Thermal cooling device sys I/F, created once it's registered:
-/sys/class/thermal/cooling_device[0-*]:
+Thermal cooling device sys I/F, created once it's registered::
+
+ /sys/class/thermal/cooling_device[0-*]:
|---type: Type of the cooling device(processor/fan/...)
|---max_state: Maximum cooling state of the cooling device
|---cur_state: Current cooling state of the cooling device
@@ -299,11 +414,13 @@ the relationship between a thermal zone and its associated cooling device.
They are created/removed for each successful execution of
thermal_zone_bind_cooling_device/thermal_zone_unbind_cooling_device.
-/sys/class/thermal/thermal_zone[0-*]:
+::
+
+ /sys/class/thermal/thermal_zone[0-*]:
|---cdev[0-*]: [0-*]th cooling device in current thermal zone
|---cdev[0-*]_trip_point: Trip point that cdev[0-*] is associated with
|---cdev[0-*]_weight: Influence of the cooling device in
- this thermal zone
+ this thermal zone
Besides the thermal zone device sysfs I/F and cooling device sysfs I/F,
the generic thermal driver also creates a hwmon sysfs I/F for each _type_
@@ -311,16 +428,17 @@ of thermal zone device. E.g. the generic thermal driver registers one hwmon
class device and build the associated hwmon sysfs I/F for all the registered
ACPI thermal zones.
-/sys/class/hwmon/hwmon[0-*]:
+::
+
+ /sys/class/hwmon/hwmon[0-*]:
|---name: The type of the thermal zone devices
|---temp[1-*]_input: The current temperature of thermal zone [1-*]
|---temp[1-*]_critical: The critical trip point of thermal zone [1-*]
Please read Documentation/hwmon/sysfs-interface.rst for additional information.
-***************************
-* Thermal zone attributes *
-***************************
+Thermal zone attributes
+-----------------------
type
Strings which represent the thermal zone type.
@@ -340,54 +458,67 @@ mode
This file gives information about the algorithm that is currently
managing the thermal zone. It can be either default kernel based
algorithm or user space application.
- enabled = enable Kernel Thermal management.
- disabled = Preventing kernel thermal zone driver actions upon
+
+ enabled
+ enable Kernel Thermal management.
+ disabled
+ Preventing kernel thermal zone driver actions upon
trip points so that user application can take full
charge of the thermal management.
+
RW, Optional
policy
One of the various thermal governors used for a particular zone.
+
RW, Required
available_policies
Available thermal governors which can be used for a particular zone.
+
RO, Required
-trip_point_[0-*]_temp
+`trip_point_[0-*]_temp`
The temperature above which trip point will be fired.
+
Unit: millidegree Celsius
+
RO, Optional
-trip_point_[0-*]_type
+`trip_point_[0-*]_type`
Strings which indicate the type of the trip point.
- E.g. it can be one of critical, hot, passive, active[0-*] for ACPI
+
+ E.g. it can be one of critical, hot, passive, `active[0-*]` for ACPI
thermal zone.
+
RO, Optional
-trip_point_[0-*]_hyst
+`trip_point_[0-*]_hyst`
The hysteresis value for a trip point, represented as an integer
Unit: Celsius
RW, Optional
-cdev[0-*]
+`cdev[0-*]`
Sysfs link to the thermal cooling device node where the sys I/F
for cooling device throttling control represents.
+
RO, Optional
-cdev[0-*]_trip_point
- The trip point in this thermal zone which cdev[0-*] is associated
+`cdev[0-*]_trip_point`
+ The trip point in this thermal zone which `cdev[0-*]` is associated
with; -1 means the cooling device is not associated with any trip
point.
+
RO, Optional
-cdev[0-*]_weight
- The influence of cdev[0-*] in this thermal zone. This value
- is relative to the rest of cooling devices in the thermal
- zone. For example, if a cooling device has a weight double
- than that of other, it's twice as effective in cooling the
- thermal zone.
- RW, Optional
+`cdev[0-*]_weight`
+ The influence of `cdev[0-*]` in this thermal zone. This value
+ is relative to the rest of cooling devices in the thermal
+ zone. For example, if a cooling device has a weight double
+ than that of other, it's twice as effective in cooling the
+ thermal zone.
+
+ RW, Optional
passive
Attribute is only present for zones in which the passive cooling
@@ -395,8 +526,11 @@ passive
and can be set to a temperature (in millidegrees) to enable a
passive trip point for the zone. Activation is done by polling with
an interval of 1 second.
+
Unit: millidegrees Celsius
+
Valid values: 0 (disabled) or greater than 1000
+
RW, Optional
emul_temp
@@ -407,17 +541,21 @@ emul_temp
threshold and its associated cooling action. This is write only node
and writing 0 on this node should disable emulation.
Unit: millidegree Celsius
+
WO, Optional
- WARNING: Be careful while enabling this option on production systems,
- because userland can easily disable the thermal policy by simply
- flooding this sysfs node with low temperature values.
+ WARNING:
+ Be careful while enabling this option on production systems,
+ because userland can easily disable the thermal policy by simply
+ flooding this sysfs node with low temperature values.
sustainable_power
An estimate of the sustained power that can be dissipated by
the thermal zone. Used by the power allocator governor. For
- more information see Documentation/thermal/power_allocator.txt
+ more information see Documentation/thermal/power_allocator.rst
+
Unit: milliwatts
+
RW, Optional
k_po
@@ -425,7 +563,8 @@ k_po
controller during temperature overshoot. Temperature overshoot
is when the current temperature is above the "desired
temperature" trip point. For more information see
- Documentation/thermal/power_allocator.txt
+ Documentation/thermal/power_allocator.rst
+
RW, Optional
k_pu
@@ -433,20 +572,23 @@ k_pu
controller during temperature undershoot. Temperature undershoot
is when the current temperature is below the "desired
temperature" trip point. For more information see
- Documentation/thermal/power_allocator.txt
+ Documentation/thermal/power_allocator.rst
+
RW, Optional
k_i
The integral term of the power allocator governor's PID
controller. This term allows the PID controller to compensate
for long term drift. For more information see
- Documentation/thermal/power_allocator.txt
+ Documentation/thermal/power_allocator.rst
+
RW, Optional
k_d
The derivative term of the power allocator governor's PID
controller. For more information see
- Documentation/thermal/power_allocator.txt
+ Documentation/thermal/power_allocator.rst
+
RW, Optional
integral_cutoff
@@ -456,8 +598,10 @@ integral_cutoff
example, if integral_cutoff is 0, then the integral term only
accumulates error when temperature is above the desired
temperature trip point. For more information see
- Documentation/thermal/power_allocator.txt
+ Documentation/thermal/power_allocator.rst
+
Unit: millidegree Celsius
+
RW, Optional
slope
@@ -465,6 +609,7 @@ slope
to determine a hotspot temperature based off the sensor's
raw readings. It is up to the device driver to determine
the usage of these values.
+
RW, Optional
offset
@@ -472,28 +617,33 @@ offset
to determine a hotspot temperature based off the sensor's
raw readings. It is up to the device driver to determine
the usage of these values.
+
RW, Optional
-*****************************
-* Cooling device attributes *
-*****************************
+Cooling device attributes
+-------------------------
type
String which represents the type of device, e.g:
+
- for generic ACPI: should be "Fan", "Processor" or "LCD"
- for memory controller device on intel_menlow platform:
should be "Memory controller".
+
RO, Required
max_state
The maximum permissible cooling state of this cooling device.
+
RO, Required
cur_state
The current cooling state of this cooling device.
The value can any integer numbers between 0 and max_state:
+
- cur_state == 0 means no cooling
- cur_state == max_state means the maximum cooling.
+
RW, Required
stats/reset
@@ -508,9 +658,11 @@ stats/time_in_state_ms:
units here is 10mS (similar to other time exported in /proc).
RO, Required
+
stats/total_trans:
A single positive value showing the total number of times the state of a
cooling device is changed.
+
RO, Required
stats/trans_table:
@@ -522,6 +674,7 @@ stats/trans_table:
RO, Required
3. A simple implementation
+==========================
ACPI thermal zone may support multiple trip points like critical, hot,
passive, active. If an ACPI thermal zone supports critical, passive,
@@ -532,11 +685,10 @@ thermal_cooling_device. Both are considered to have the same
effectiveness in cooling the thermal zone.
If the processor is listed in _PSL method, and the fan is listed in _AL0
-method, the sys I/F structure will be built like this:
+method, the sys I/F structure will be built like this::
-/sys/class/thermal:
-
-|thermal_zone1:
+ /sys/class/thermal:
+ |thermal_zone1:
|---type: acpitz
|---temp: 37000
|---mode: enabled
@@ -557,24 +709,24 @@ method, the sys I/F structure will be built like this:
|---cdev1_trip_point: 2 /* cdev1 can be used for active[0]*/
|---cdev1_weight: 1024
-|cooling_device0:
+ |cooling_device0:
|---type: Processor
|---max_state: 8
|---cur_state: 0
-|cooling_device3:
+ |cooling_device3:
|---type: Fan
|---max_state: 2
|---cur_state: 0
-/sys/class/hwmon:
-
-|hwmon0:
+ /sys/class/hwmon:
+ |hwmon0:
|---name: acpitz
|---temp1_input: 37000
|---temp1_crit: 100000
4. Event Notification
+=====================
The framework includes a simple notification mechanism, in the form of a
netlink event. Netlink socket initialization is done during the _init_
@@ -587,21 +739,28 @@ event will be one of:{THERMAL_AUX0, THERMAL_AUX1, THERMAL_CRITICAL,
THERMAL_DEV_FAULT}. Notification can be sent when the current temperature
crosses any of the configured thresholds.
-5. Export Symbol APIs:
+5. Export Symbol APIs
+=====================
+
+5.1. get_tz_trend
+-----------------
-5.1: get_tz_trend:
This function returns the trend of a thermal zone, i.e the rate of change
of temperature of the thermal zone. Ideally, the thermal sensor drivers
are supposed to implement the callback. If they don't, the thermal
framework calculated the trend by comparing the previous and the current
temperature values.
-5.2:get_thermal_instance:
+5.2. get_thermal_instance
+-------------------------
+
This function returns the thermal_instance corresponding to a given
{thermal_zone, cooling_device, trip_point} combination. Returns NULL
if such an instance does not exist.
-5.3:thermal_notify_framework:
+5.3. thermal_notify_framework
+-----------------------------
+
This function handles the trip events from sensor drivers. It starts
throttling the cooling devices according to the policy configured.
For CRITICAL and HOT trip points, this notifies the respective drivers,
@@ -609,12 +768,15 @@ and does actual throttling for other trip points i.e ACTIVE and PASSIVE.
The throttling policy is based on the configured platform data; if no
platform data is provided, this uses the step_wise throttling policy.
-5.4:thermal_cdev_update:
+5.4. thermal_cdev_update
+------------------------
+
This function serves as an arbitrator to set the state of a cooling
device. It sets the cooling device to the deepest cooling state if
possible.
-6. thermal_emergency_poweroff:
+6. thermal_emergency_poweroff
+=============================
On an event of critical trip temperature crossing. Thermal framework
allows the system to shutdown gracefully by calling orderly_poweroff().
diff --git a/Documentation/thermal/x86_pkg_temperature_thermal b/Documentation/thermal/x86_pkg_temperature_thermal.rst
similarity index 80%
rename from Documentation/thermal/x86_pkg_temperature_thermal
rename to Documentation/thermal/x86_pkg_temperature_thermal.rst
index 17a3a4c0a0ca..f134dbd3f5a9 100644
--- a/Documentation/thermal/x86_pkg_temperature_thermal
+++ b/Documentation/thermal/x86_pkg_temperature_thermal.rst
@@ -1,19 +1,23 @@
+===================================
Kernel driver: x86_pkg_temp_thermal
-===================
+===================================
Supported chips:
+
* x86: with package level thermal management
+
(Verify using: CPUID.06H:EAX[bit 6] =1)
Authors: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Reference
----
+---------
+
Intel® 64 and IA-32 Architectures Software Developer’s Manual (Jan, 2013):
Chapter 14.6: PACKAGE LEVEL THERMAL MANAGEMENT
Description
----------
+-----------
This driver register CPU digital temperature package level sensor as a thermal
zone with maximum two user mode configurable trip points. Number of trip points
@@ -25,23 +29,27 @@ take any action to control temperature.
Threshold management
--------------------
Each package will register as a thermal zone under /sys/class/thermal.
-Example:
-/sys/class/thermal/thermal_zone1
+
+Example::
+
+ /sys/class/thermal/thermal_zone1
This contains two trip points:
+
- trip_point_0_temp
- trip_point_1_temp
User can set any temperature between 0 to TJ-Max temperature. Temperature units
-are in milli-degree Celsius. Refer to "Documentation/thermal/sysfs-api.txt" for
+are in milli-degree Celsius. Refer to "Documentation/thermal/sysfs-api.rst" for
thermal sys-fs details.
Any value other than 0 in these trip points, can trigger thermal notifications.
Setting 0, stops sending thermal notifications.
-Thermal notifications: To get kobject-uevent notifications, set the thermal zone
-policy to "user_space". For example: echo -n "user_space" > policy
-
-
+Thermal notifications:
+To get kobject-uevent notifications, set the thermal zone
+policy to "user_space".
+For example::
+ echo -n "user_space" > policy
diff --git a/MAINTAINERS b/MAINTAINERS
index 8aa3d0bd5b80..85648a944446 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -15762,7 +15762,7 @@ M: Viresh Kumar <viresh.kumar@linaro.org>
M: Javi Merino <javi.merino@kernel.org>
L: linux-pm@vger.kernel.org
S: Supported
-F: Documentation/thermal/cpu-cooling-api.txt
+F: Documentation/thermal/cpu-cooling-api.rst
F: drivers/thermal/cpu_cooling.c
F: include/linux/cpu_cooling.h
diff --git a/include/linux/thermal.h b/include/linux/thermal.h
index 15a4ca5d7099..681047f8cc05 100644
--- a/include/linux/thermal.h
+++ b/include/linux/thermal.h
@@ -251,7 +251,7 @@ struct thermal_bind_params {
* platform characterization. This value is relative to the
* rest of the weights so a cooling device whose weight is
* double that of another cooling device is twice as
- * effective. See Documentation/thermal/sysfs-api.txt for more
+ * effective. See Documentation/thermal/sysfs-api.rst for more
* information.
*/
int weight;
@@ -259,7 +259,7 @@ struct thermal_bind_params {
/*
* This is a bit mask that gives the binding relation between this
* thermal zone and cdev, for a particular trip point.
- * See Documentation/thermal/sysfs-api.txt for more information.
+ * See Documentation/thermal/sysfs-api.rst for more information.
*/
int trip_mask;
--
2.21.0
^ permalink raw reply related
* Re: [PATCH V4 2/3] thermal/drivers/cpu_cooling: Unregister with the policy
From: Daniel Lezcano @ 2019-06-28 11:57 UTC (permalink / raw)
To: Rafael J. Wysocki
Cc: Viresh Kumar, Rafael J. Wysocki, Eduardo Valentin,
Linux Kernel Mailing List, Sudeep Holla, Amit Daniel Kachhap,
Javi Merino, Zhang Rui, Shawn Guo, Sascha Hauer,
Pengutronix Kernel Team, Fabio Estevam, NXP Linux Team, Keerthy,
open list:CPU FREQUENCY DRIVERS - ARM BIG LITTLE,
moderated list:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE,
open list:TI BANDGAP AND THERMAL DRIVER, Quentin Perret
In-Reply-To: <CAJZ5v0hPjPMLfiWMPnm59HpmTbJNXw0MDpvRwMGOUpMXYUwgRw@mail.gmail.com>
On 28/06/2019 12:03, Rafael J. Wysocki wrote:
> On Fri, Jun 28, 2019 at 11:58 AM Daniel Lezcano
> <daniel.lezcano@linaro.org> wrote:
>>
>>
>> On 28/06/2019 11:12, Rafael J. Wysocki wrote:
>>> On Thu, Jun 27, 2019 at 11:02 PM Daniel Lezcano
>>> <daniel.lezcano@linaro.org> wrote:
>>>>
>>>> Currently the function cpufreq_cooling_register() returns a cooling
>>>> device pointer which is used back as a pointer to call the function
>>>> cpufreq_cooling_unregister(). Even if it is correct, it would make
>>>> sense to not leak the structure inside a cpufreq driver and keep the
>>>> code thermal code self-encapsulate. Moreover, that forces to add an
>>>> extra variable in each driver using this function.
>>>>
>>>> Instead of passing the cooling device to unregister, pass the policy.
>>>>
>>>> Because the cpufreq_cooling_unregister() function uses the policy to
>>>> unregister itself. The only purpose of the cooling device pointer is
>>>> to unregister the cpu cooling device.
>>>>
>>>> As there is no more need of this pointer, remove it.
>>>>
>>>> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
>>>> Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
>>>
>>> This doesn't apply for me.
>>>
>>> Care to rebase it on top of the Linus' tree?
>>
>> Sure but the patch depends on 1/3 which is in bleeding edge. Shall I
>> rebase the 3 patches on v5.2-rc6 and resend ?
>
> You can do that.
>
> Alternatively, you can rebase on top of my linux-next branch.
Ok, it is rebased on top of linux-next, however the conflict is coming
from the energy model patchset sent by Quentin [1][2] I used to based my
series which is not yet applied in the thermal tree.
I'm wondering if it wouldn't make sense to take Quentin's series also,
it is a long time around in the mailing list, reviewed and acked. So I
can send the two remaining patches on top of it without conflict,
otherwise we will have a conflict in the merge window.
[1] https://lkml.org/lkml/2019/5/30/794
[2] https://lkml.org/lkml/2019/6/19/190
--
<http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs
Follow Linaro: <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog
^ permalink raw reply
* Re: [PATCH] devfreq: tegra20: add COMMON_CLK dependency
From: Dmitry Osipenko @ 2019-06-28 11:05 UTC (permalink / raw)
To: Arnd Bergmann, MyungJoo Ham, Kyungmin Park
Cc: Chanwoo Choi, linux-pm, linux-kernel
In-Reply-To: <c2701295-0fd6-c5ec-59d4-6e6b3b3bdb8e@gmail.com>
28.06.2019 14:00, Dmitry Osipenko пишет:
> 28.06.2019 13:32, Arnd Bergmann пишет:
>> Compile-testing the new driver on platforms without CONFIG_COMMON_CLK
>> leads to a link error:
>>
>> drivers/devfreq/tegra20-devfreq.o: In function `tegra_devfreq_target':
>> tegra20-devfreq.c:(.text+0x288): undefined reference to `clk_set_min_rate'
>>
>> Add a dependency on COMMON_CLK to avoid this.
>>
>> Fixes: 1d39ee8dad6d ("PM / devfreq: Introduce driver for NVIDIA Tegra20")
>> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
>> ---
>> drivers/devfreq/Kconfig | 1 +
>> 1 file changed, 1 insertion(+)
>>
>> diff --git a/drivers/devfreq/Kconfig b/drivers/devfreq/Kconfig
>> index f3b242987fd9..defe1d438710 100644
>> --- a/drivers/devfreq/Kconfig
>> +++ b/drivers/devfreq/Kconfig
>> @@ -107,6 +107,7 @@ config ARM_TEGRA_DEVFREQ
>> config ARM_TEGRA20_DEVFREQ
>> tristate "NVIDIA Tegra20 DEVFREQ Driver"
>> depends on (TEGRA_MC && TEGRA20_EMC) || COMPILE_TEST
>> + depends on COMMON_CLK
>> select DEVFREQ_GOV_SIMPLE_ONDEMAND
>> select PM_OPP
>> help
>>
>
> How is it possible to happen? There is a stub for clk_set_min_rate() when COMMON_CLK
> is disabled .. Could you please show the kernel's config that causes the problem?
>
> Moreover that was me who added the missing stub for clk_set_min_rate() in [1] to fix [2].
>
> [1]
> https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/include/linux/clk.h?id=b88c9f4129dcec941e5a26508e991c08051ed1ac
>
> [2] https://lists.01.org/pipermail/kbuild-all/2019-April/060292.html
>
> So this patch looks wrong to me because apparently it fixes a non-existent problem.
>
Oh, wait. That was "implicit declaration" and this one is "undefined reference"!
Still looks like something else need to be fixed..
^ permalink raw reply
* Re: [PATCH] devfreq: tegra20: add COMMON_CLK dependency
From: Dmitry Osipenko @ 2019-06-28 11:00 UTC (permalink / raw)
To: Arnd Bergmann, MyungJoo Ham, Kyungmin Park
Cc: Chanwoo Choi, linux-pm, linux-kernel
In-Reply-To: <20190628103232.2467959-1-arnd@arndb.de>
28.06.2019 13:32, Arnd Bergmann пишет:
> Compile-testing the new driver on platforms without CONFIG_COMMON_CLK
> leads to a link error:
>
> drivers/devfreq/tegra20-devfreq.o: In function `tegra_devfreq_target':
> tegra20-devfreq.c:(.text+0x288): undefined reference to `clk_set_min_rate'
>
> Add a dependency on COMMON_CLK to avoid this.
>
> Fixes: 1d39ee8dad6d ("PM / devfreq: Introduce driver for NVIDIA Tegra20")
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
> drivers/devfreq/Kconfig | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/drivers/devfreq/Kconfig b/drivers/devfreq/Kconfig
> index f3b242987fd9..defe1d438710 100644
> --- a/drivers/devfreq/Kconfig
> +++ b/drivers/devfreq/Kconfig
> @@ -107,6 +107,7 @@ config ARM_TEGRA_DEVFREQ
> config ARM_TEGRA20_DEVFREQ
> tristate "NVIDIA Tegra20 DEVFREQ Driver"
> depends on (TEGRA_MC && TEGRA20_EMC) || COMPILE_TEST
> + depends on COMMON_CLK
> select DEVFREQ_GOV_SIMPLE_ONDEMAND
> select PM_OPP
> help
>
How is it possible to happen? There is a stub for clk_set_min_rate() when COMMON_CLK
is disabled .. Could you please show the kernel's config that causes the problem?
Moreover that was me who added the missing stub for clk_set_min_rate() in [1] to fix [2].
[1]
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/include/linux/clk.h?id=b88c9f4129dcec941e5a26508e991c08051ed1ac
[2] https://lists.01.org/pipermail/kbuild-all/2019-April/060292.html
So this patch looks wrong to me because apparently it fixes a non-existent problem.
^ permalink raw reply
* [PATCH] devfreq: tegra20: add COMMON_CLK dependency
From: Arnd Bergmann @ 2019-06-28 10:32 UTC (permalink / raw)
To: MyungJoo Ham, Kyungmin Park
Cc: Arnd Bergmann, Chanwoo Choi, Dmitry Osipenko, linux-pm,
linux-kernel
Compile-testing the new driver on platforms without CONFIG_COMMON_CLK
leads to a link error:
drivers/devfreq/tegra20-devfreq.o: In function `tegra_devfreq_target':
tegra20-devfreq.c:(.text+0x288): undefined reference to `clk_set_min_rate'
Add a dependency on COMMON_CLK to avoid this.
Fixes: 1d39ee8dad6d ("PM / devfreq: Introduce driver for NVIDIA Tegra20")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
drivers/devfreq/Kconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/devfreq/Kconfig b/drivers/devfreq/Kconfig
index f3b242987fd9..defe1d438710 100644
--- a/drivers/devfreq/Kconfig
+++ b/drivers/devfreq/Kconfig
@@ -107,6 +107,7 @@ config ARM_TEGRA_DEVFREQ
config ARM_TEGRA20_DEVFREQ
tristate "NVIDIA Tegra20 DEVFREQ Driver"
depends on (TEGRA_MC && TEGRA20_EMC) || COMPILE_TEST
+ depends on COMMON_CLK
select DEVFREQ_GOV_SIMPLE_ONDEMAND
select PM_OPP
help
--
2.20.0
^ permalink raw reply related
* Re: [PATCH v2 25/27] kernel: power: use kzalloc rather than kmalloc followed with memset
From: Rafael J. Wysocki @ 2019-06-28 10:16 UTC (permalink / raw)
To: Fuqian Huang; +Cc: Len Brown, Pavel Machek, linux-pm, linux-kernel
In-Reply-To: <20190628025036.16135-1-huangfq.daxian@gmail.com>
On Friday, June 28, 2019 4:50:35 AM CEST Fuqian Huang wrote:
> Use zeroing allocator instead of using allocator
> followed with memset with 0
>
> Signed-off-by: Fuqian Huang <huangfq.daxian@gmail.com>
> ---
> kernel/power/swap.c | 3 +--
> 1 file changed, 1 insertion(+), 2 deletions(-)
>
> diff --git a/kernel/power/swap.c b/kernel/power/swap.c
> index e1912ad13bdc..ca0fcb5ced71 100644
> --- a/kernel/power/swap.c
> +++ b/kernel/power/swap.c
> @@ -974,12 +974,11 @@ static int get_swap_reader(struct swap_map_handle *handle,
> last = handle->maps = NULL;
> offset = swsusp_header->image;
> while (offset) {
> - tmp = kmalloc(sizeof(*handle->maps), GFP_KERNEL);
> + tmp = kzalloc(sizeof(*handle->maps), GFP_KERNEL);
> if (!tmp) {
> release_swap_reader(handle);
> return -ENOMEM;
> }
> - memset(tmp, 0, sizeof(*tmp));
> if (!handle->maps)
> handle->maps = tmp;
> if (last)
>
Queued up for 5.3 with minor subject changes. Thanks!
^ permalink raw reply
* Re: [PATCH V4 2/3] thermal/drivers/cpu_cooling: Unregister with the policy
From: Rafael J. Wysocki @ 2019-06-28 10:07 UTC (permalink / raw)
To: Daniel Lezcano
Cc: Viresh Kumar, Rafael J. Wysocki, Eduardo Valentin,
Linux Kernel Mailing List, Sudeep Holla, Amit Daniel Kachhap,
Javi Merino, Zhang Rui, Shawn Guo, Sascha Hauer,
Pengutronix Kernel Team, Fabio Estevam, NXP Linux Team, Keerthy,
open list:CPU FREQUENCY DRIVERS - ARM BIG LITTLE,
moderated list:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE,
open list:TI BANDGAP AND THERMAL DRIVER
In-Reply-To: <CAJZ5v0jJzCGfQFqi-S3vqs74D73MaE4f7WYF_NVnDKawNV4Wzw@mail.gmail.com>
On Fri, Jun 28, 2019 at 11:12 AM Rafael J. Wysocki <rafael@kernel.org> wrote:
>
> On Thu, Jun 27, 2019 at 11:02 PM Daniel Lezcano
> <daniel.lezcano@linaro.org> wrote:
> >
> > Currently the function cpufreq_cooling_register() returns a cooling
> > device pointer which is used back as a pointer to call the function
> > cpufreq_cooling_unregister(). Even if it is correct, it would make
> > sense to not leak the structure inside a cpufreq driver and keep the
> > code thermal code self-encapsulate. Moreover, that forces to add an
> > extra variable in each driver using this function.
> >
> > Instead of passing the cooling device to unregister, pass the policy.
> >
> > Because the cpufreq_cooling_unregister() function uses the policy to
> > unregister itself. The only purpose of the cooling device pointer is
> > to unregister the cpu cooling device.
> >
> > As there is no more need of this pointer, remove it.
> >
> > Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
> > Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
[cut]
> > -void cpufreq_cooling_unregister(struct thermal_cooling_device *cdev)
> > +void cpufreq_cooling_unregister(struct cpufreq_policy *policy)
> > {
> > struct cpufreq_cooling_device *cpufreq_cdev;
>
> I would do
>
> struct cpufreq_cooling_device *ccd, *cpufreq_cdev = NULL;
>
> and then ->
Not even that. ->
>
> > bool last;
> >
> > - if (!cdev)
> > - return;
> > -
> > - cpufreq_cdev = cdev->devdata;
> > -
> > mutex_lock(&cooling_list_lock);
> > - list_del(&cpufreq_cdev->node);
> > - /* Unregister the notifier for the last cpufreq cooling device */
> > - last = list_empty(&cpufreq_cdev_list);
> > + list_for_each_entry(cpufreq_cdev, &cpufreq_cdev_list, node) {
>
> -> list_for_each_entry(ccd, &cpufreq_cdev_list, node) {
> if (ccd->policy == policy) {
>
> > + if (cpufreq_cdev->policy == policy) {
>
> cpufreq_cdev = ccd;
>
> > + list_del(&cpufreq_cdev->node);
> > + last = list_empty(&cpufreq_cdev_list);
> > + break;
> > + }
> > + }
> > mutex_unlock(&cooling_list_lock);
>
> And here
>
> if (!cpufreq_cdev)
> return;
-> It would be sufficient to simply do:
if (cpufreq_cdev->policy != policy)
return;
here AFAICS.
>
> And that's it. No new functions needed.
>
> > - if (last)
> > - cpufreq_unregister_notifier(&thermal_cpufreq_notifier_block,
> > - CPUFREQ_POLICY_NOTIFIER);
> > -
>
> And I don't that the above needs to be changed at all in any case.
>
>
> > - thermal_cooling_device_unregister(cdev);
> > - ida_simple_remove(&cpufreq_ida, cpufreq_cdev->id);
> > - kfree(cpufreq_cdev->idle_time);
> > - kfree(cpufreq_cdev);
> > + if (cpufreq_cdev->policy == policy)
> > + __cpufreq_cooling_unregister(cpufreq_cdev, last);
> > }
^ permalink raw reply
* Re: [PATCH V4 2/3] thermal/drivers/cpu_cooling: Unregister with the policy
From: Rafael J. Wysocki @ 2019-06-28 10:03 UTC (permalink / raw)
To: Daniel Lezcano
Cc: Rafael J. Wysocki, Viresh Kumar, Rafael J. Wysocki,
Eduardo Valentin, Linux Kernel Mailing List, Sudeep Holla,
Amit Daniel Kachhap, Javi Merino, Zhang Rui, Shawn Guo,
Sascha Hauer, Pengutronix Kernel Team, Fabio Estevam,
NXP Linux Team, Keerthy,
open list:CPU FREQUENCY DRIVERS - ARM BIG LITTLE,
moderated list:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE,
open list:TI BANDGAP AND THERMAL DRIVER
In-Reply-To: <cd79ffe3-4749-9135-27c4-deb4f259d702@linaro.org>
On Fri, Jun 28, 2019 at 11:58 AM Daniel Lezcano
<daniel.lezcano@linaro.org> wrote:
>
>
> On 28/06/2019 11:12, Rafael J. Wysocki wrote:
> > On Thu, Jun 27, 2019 at 11:02 PM Daniel Lezcano
> > <daniel.lezcano@linaro.org> wrote:
> >>
> >> Currently the function cpufreq_cooling_register() returns a cooling
> >> device pointer which is used back as a pointer to call the function
> >> cpufreq_cooling_unregister(). Even if it is correct, it would make
> >> sense to not leak the structure inside a cpufreq driver and keep the
> >> code thermal code self-encapsulate. Moreover, that forces to add an
> >> extra variable in each driver using this function.
> >>
> >> Instead of passing the cooling device to unregister, pass the policy.
> >>
> >> Because the cpufreq_cooling_unregister() function uses the policy to
> >> unregister itself. The only purpose of the cooling device pointer is
> >> to unregister the cpu cooling device.
> >>
> >> As there is no more need of this pointer, remove it.
> >>
> >> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
> >> Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
> >
> > This doesn't apply for me.
> >
> > Care to rebase it on top of the Linus' tree?
>
> Sure but the patch depends on 1/3 which is in bleeding edge. Shall I
> rebase the 3 patches on v5.2-rc6 and resend ?
You can do that.
Alternatively, you can rebase on top of my linux-next branch.
^ permalink raw reply
* Re: [PATCH V4 2/3] thermal/drivers/cpu_cooling: Unregister with the policy
From: Daniel Lezcano @ 2019-06-28 9:57 UTC (permalink / raw)
To: Rafael J. Wysocki
Cc: Viresh Kumar, Rafael J. Wysocki, Eduardo Valentin,
Linux Kernel Mailing List, Sudeep Holla, Amit Daniel Kachhap,
Javi Merino, Zhang Rui, Shawn Guo, Sascha Hauer,
Pengutronix Kernel Team, Fabio Estevam, NXP Linux Team, Keerthy,
open list:CPU FREQUENCY DRIVERS - ARM BIG LITTLE,
moderated list:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE,
open list:TI BANDGAP AND THERMAL DRIVER
In-Reply-To: <CAJZ5v0jJzCGfQFqi-S3vqs74D73MaE4f7WYF_NVnDKawNV4Wzw@mail.gmail.com>
On 28/06/2019 11:12, Rafael J. Wysocki wrote:
> On Thu, Jun 27, 2019 at 11:02 PM Daniel Lezcano
> <daniel.lezcano@linaro.org> wrote:
>>
>> Currently the function cpufreq_cooling_register() returns a cooling
>> device pointer which is used back as a pointer to call the function
>> cpufreq_cooling_unregister(). Even if it is correct, it would make
>> sense to not leak the structure inside a cpufreq driver and keep the
>> code thermal code self-encapsulate. Moreover, that forces to add an
>> extra variable in each driver using this function.
>>
>> Instead of passing the cooling device to unregister, pass the policy.
>>
>> Because the cpufreq_cooling_unregister() function uses the policy to
>> unregister itself. The only purpose of the cooling device pointer is
>> to unregister the cpu cooling device.
>>
>> As there is no more need of this pointer, remove it.
>>
>> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
>> Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
>
> This doesn't apply for me.
>
> Care to rebase it on top of the Linus' tree?
Sure but the patch depends on 1/3 which is in bleeding edge. Shall I
rebase the 3 patches on v5.2-rc6 and resend ?
--
<http://www.linaro.org/> Linaro.org │ Open source software for ARM SoCs
Follow Linaro: <http://www.facebook.com/pages/Linaro> Facebook |
<http://twitter.com/#!/linaroorg> Twitter |
<http://www.linaro.org/linaro-blog/> Blog
^ permalink raw reply
* Re: [PATCH V4 2/3] thermal/drivers/cpu_cooling: Unregister with the policy
From: Rafael J. Wysocki @ 2019-06-28 9:12 UTC (permalink / raw)
To: Daniel Lezcano
Cc: Viresh Kumar, Rafael J. Wysocki, Eduardo Valentin,
Linux Kernel Mailing List, Sudeep Holla, Amit Daniel Kachhap,
Javi Merino, Zhang Rui, Shawn Guo, Sascha Hauer,
Pengutronix Kernel Team, Fabio Estevam, NXP Linux Team, Keerthy,
open list:CPU FREQUENCY DRIVERS - ARM BIG LITTLE,
moderated list:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE,
open list:TI BANDGAP AND THERMAL DRIVER
In-Reply-To: <20190627210209.32600-2-daniel.lezcano@linaro.org>
On Thu, Jun 27, 2019 at 11:02 PM Daniel Lezcano
<daniel.lezcano@linaro.org> wrote:
>
> Currently the function cpufreq_cooling_register() returns a cooling
> device pointer which is used back as a pointer to call the function
> cpufreq_cooling_unregister(). Even if it is correct, it would make
> sense to not leak the structure inside a cpufreq driver and keep the
> code thermal code self-encapsulate. Moreover, that forces to add an
> extra variable in each driver using this function.
>
> Instead of passing the cooling device to unregister, pass the policy.
>
> Because the cpufreq_cooling_unregister() function uses the policy to
> unregister itself. The only purpose of the cooling device pointer is
> to unregister the cpu cooling device.
>
> As there is no more need of this pointer, remove it.
>
> Signed-off-by: Daniel Lezcano <daniel.lezcano@linaro.org>
> Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
This doesn't apply for me.
Care to rebase it on top of the Linus' tree?
Also see below.
> ---
> drivers/cpufreq/arm_big_little.c | 9 ++--
> drivers/cpufreq/cpufreq.c | 8 ++--
> drivers/thermal/cpu_cooling.c | 42 +++++++++++--------
> drivers/thermal/imx_thermal.c | 12 +++---
> .../ti-soc-thermal/ti-thermal-common.c | 10 ++---
> include/linux/cpu_cooling.h | 6 +--
> include/linux/cpufreq.h | 3 --
> 7 files changed, 45 insertions(+), 45 deletions(-)
>
> diff --git a/drivers/cpufreq/arm_big_little.c b/drivers/cpufreq/arm_big_little.c
> index 7fe52fcddcf1..718c63231e66 100644
> --- a/drivers/cpufreq/arm_big_little.c
> +++ b/drivers/cpufreq/arm_big_little.c
> @@ -56,7 +56,6 @@ static bool bL_switching_enabled;
> #define ACTUAL_FREQ(cluster, freq) ((cluster == A7_CLUSTER) ? freq << 1 : freq)
> #define VIRT_FREQ(cluster, freq) ((cluster == A7_CLUSTER) ? freq >> 1 : freq)
>
> -static struct thermal_cooling_device *cdev[MAX_CLUSTERS];
> static const struct cpufreq_arm_bL_ops *arm_bL_ops;
> static struct clk *clk[MAX_CLUSTERS];
> static struct cpufreq_frequency_table *freq_table[MAX_CLUSTERS + 1];
> @@ -501,10 +500,8 @@ static int bL_cpufreq_exit(struct cpufreq_policy *policy)
> struct device *cpu_dev;
> int cur_cluster = cpu_to_cluster(policy->cpu);
>
> - if (cur_cluster < MAX_CLUSTERS) {
> - cpufreq_cooling_unregister(cdev[cur_cluster]);
> - cdev[cur_cluster] = NULL;
> - }
> + if (cur_cluster < MAX_CLUSTERS)
> + cpufreq_cooling_unregister(policy);
>
> cpu_dev = get_cpu_device(policy->cpu);
> if (!cpu_dev) {
> @@ -527,7 +524,7 @@ static void bL_cpufreq_ready(struct cpufreq_policy *policy)
> if (cur_cluster >= MAX_CLUSTERS)
> return;
>
> - cdev[cur_cluster] = of_cpufreq_cooling_register(policy);
> + of_cpufreq_cooling_register(policy);
> }
>
> static struct cpufreq_driver bL_cpufreq_driver = {
> diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c
> index aee024e42618..1663a5601811 100644
> --- a/drivers/cpufreq/cpufreq.c
> +++ b/drivers/cpufreq/cpufreq.c
> @@ -1379,7 +1379,7 @@ static int cpufreq_online(unsigned int cpu)
> cpufreq_driver->ready(policy);
>
> if (cpufreq_thermal_control_enabled(cpufreq_driver))
> - policy->cdev = of_cpufreq_cooling_register(policy);
> + of_cpufreq_cooling_register(policy);
>
> pr_debug("initialization complete\n");
>
> @@ -1468,10 +1468,8 @@ static int cpufreq_offline(unsigned int cpu)
> goto unlock;
> }
>
> - if (cpufreq_thermal_control_enabled(cpufreq_driver)) {
> - cpufreq_cooling_unregister(policy->cdev);
> - policy->cdev = NULL;
> - }
> + if (cpufreq_thermal_control_enabled(cpufreq_driver))
> + cpufreq_cooling_unregister(policy);
>
> if (cpufreq_driver->stop_cpu)
> cpufreq_driver->stop_cpu(policy);
> diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c
> index 83486775e593..be01546a656f 100644
> --- a/drivers/thermal/cpu_cooling.c
> +++ b/drivers/thermal/cpu_cooling.c
> @@ -78,6 +78,7 @@ struct cpufreq_cooling_device {
> struct cpufreq_policy *policy;
> struct list_head node;
> struct time_in_idle *idle_time;
> + struct thermal_cooling_device *cdev;
> };
>
> static DEFINE_IDA(cpufreq_ida);
> @@ -606,6 +607,7 @@ __cpufreq_cooling_register(struct device_node *np,
> goto remove_ida;
>
> cpufreq_cdev->clipped_freq = get_state_freq(cpufreq_cdev, 0);
> + cpufreq_cdev->cdev = cdev;
>
> mutex_lock(&cooling_list_lock);
> /* Register the notifier for first cpufreq cooling device */
> @@ -693,35 +695,41 @@ of_cpufreq_cooling_register(struct cpufreq_policy *policy)
> }
> EXPORT_SYMBOL_GPL(of_cpufreq_cooling_register);
>
> +void __cpufreq_cooling_unregister(struct cpufreq_cooling_device *cpufreq_cdev, int last)
> +{
> + /* Unregister the notifier for the last cpufreq cooling device */
> + if (last)
> + cpufreq_unregister_notifier(&thermal_cpufreq_notifier_block,
> + CPUFREQ_POLICY_NOTIFIER);
> +
> + thermal_cooling_device_unregister(cpufreq_cdev->cdev);
> + ida_simple_remove(&cpufreq_ida, cpufreq_cdev->id);
> + kfree(cpufreq_cdev->idle_time);
> + kfree(cpufreq_cdev);
> +}
> +
> /**
> * cpufreq_cooling_unregister - function to remove cpufreq cooling device.
> * @cdev: thermal cooling device pointer.
> *
> * This interface function unregisters the "thermal-cpufreq-%x" cooling device.
> */
> -void cpufreq_cooling_unregister(struct thermal_cooling_device *cdev)
> +void cpufreq_cooling_unregister(struct cpufreq_policy *policy)
> {
> struct cpufreq_cooling_device *cpufreq_cdev;
I would do
struct cpufreq_cooling_device *ccd, *cpufreq_cdev = NULL;
and then ->
> bool last;
>
> - if (!cdev)
> - return;
> -
> - cpufreq_cdev = cdev->devdata;
> -
> mutex_lock(&cooling_list_lock);
> - list_del(&cpufreq_cdev->node);
> - /* Unregister the notifier for the last cpufreq cooling device */
> - last = list_empty(&cpufreq_cdev_list);
> + list_for_each_entry(cpufreq_cdev, &cpufreq_cdev_list, node) {
-> list_for_each_entry(ccd, &cpufreq_cdev_list, node) {
if (ccd->policy == policy) {
> + if (cpufreq_cdev->policy == policy) {
cpufreq_cdev = ccd;
> + list_del(&cpufreq_cdev->node);
> + last = list_empty(&cpufreq_cdev_list);
> + break;
> + }
> + }
> mutex_unlock(&cooling_list_lock);
And here
if (!cpufreq_cdev)
return;
And that's it. No new functions needed.
> - if (last)
> - cpufreq_unregister_notifier(&thermal_cpufreq_notifier_block,
> - CPUFREQ_POLICY_NOTIFIER);
> -
And I don't that the above needs to be changed at all in any case.
> - thermal_cooling_device_unregister(cdev);
> - ida_simple_remove(&cpufreq_ida, cpufreq_cdev->id);
> - kfree(cpufreq_cdev->idle_time);
> - kfree(cpufreq_cdev);
> + if (cpufreq_cdev->policy == policy)
> + __cpufreq_cooling_unregister(cpufreq_cdev, last);
> }
> EXPORT_SYMBOL_GPL(cpufreq_cooling_unregister);
> diff --git a/drivers/thermal/imx_thermal.c b/drivers/thermal/imx_thermal.c
> index bb6754a5342c..021c0948b740 100644
> --- a/drivers/thermal/imx_thermal.c
> +++ b/drivers/thermal/imx_thermal.c
> @@ -203,7 +203,6 @@ static struct thermal_soc_data thermal_imx7d_data = {
> struct imx_thermal_data {
> struct cpufreq_policy *policy;
> struct thermal_zone_device *tz;
> - struct thermal_cooling_device *cdev;
> enum thermal_device_mode mode;
> struct regmap *tempmon;
> u32 c1, c2; /* See formula in imx_init_calib() */
> @@ -656,6 +655,7 @@ MODULE_DEVICE_TABLE(of, of_imx_thermal_match);
> static int imx_thermal_register_legacy_cooling(struct imx_thermal_data *data)
> {
> struct device_node *np;
> + struct thermal_cooling_device *cdev;
> int ret;
>
> data->policy = cpufreq_cpu_get(0);
> @@ -667,9 +667,9 @@ static int imx_thermal_register_legacy_cooling(struct imx_thermal_data *data)
> np = of_get_cpu_node(data->policy->cpu, NULL);
>
> if (!np || !of_find_property(np, "#cooling-cells", NULL)) {
> - data->cdev = cpufreq_cooling_register(data->policy);
> - if (IS_ERR(data->cdev)) {
> - ret = PTR_ERR(data->cdev);
> + cdev = cpufreq_cooling_register(data->policy);
> + if (IS_ERR(cdev)) {
> + ret = PTR_ERR(cdev);
> cpufreq_cpu_put(data->policy);
> return ret;
> }
> @@ -680,7 +680,7 @@ static int imx_thermal_register_legacy_cooling(struct imx_thermal_data *data)
>
> static void imx_thermal_unregister_legacy_cooling(struct imx_thermal_data *data)
> {
> - cpufreq_cooling_unregister(data->cdev);
> + cpufreq_cooling_unregister(data->policy);
> cpufreq_cpu_put(data->policy);
> }
>
> @@ -872,7 +872,7 @@ static int imx_thermal_remove(struct platform_device *pdev)
> clk_disable_unprepare(data->thermal_clk);
>
> thermal_zone_device_unregister(data->tz);
> - cpufreq_cooling_unregister(data->cdev);
> + cpufreq_cooling_unregister(data->policy);
> cpufreq_cpu_put(data->policy);
>
> return 0;
> diff --git a/drivers/thermal/ti-soc-thermal/ti-thermal-common.c b/drivers/thermal/ti-soc-thermal/ti-thermal-common.c
> index b4f981daeaf2..170b70b6ec61 100644
> --- a/drivers/thermal/ti-soc-thermal/ti-thermal-common.c
> +++ b/drivers/thermal/ti-soc-thermal/ti-thermal-common.c
> @@ -41,7 +41,6 @@ struct ti_thermal_data {
> struct cpufreq_policy *policy;
> struct thermal_zone_device *ti_thermal;
> struct thermal_zone_device *pcb_tz;
> - struct thermal_cooling_device *cool_dev;
> struct ti_bandgap *bgp;
> enum thermal_device_mode mode;
> struct work_struct thermal_wq;
> @@ -233,6 +232,7 @@ int ti_thermal_register_cpu_cooling(struct ti_bandgap *bgp, int id)
> {
> struct ti_thermal_data *data;
> struct device_node *np = bgp->dev->of_node;
> + struct thermal_cooling_device *cdev;
>
> /*
> * We are assuming here that if one deploys the zone
> @@ -256,9 +256,9 @@ int ti_thermal_register_cpu_cooling(struct ti_bandgap *bgp, int id)
> }
>
> /* Register cooling device */
> - data->cool_dev = cpufreq_cooling_register(data->policy);
> - if (IS_ERR(data->cool_dev)) {
> - int ret = PTR_ERR(data->cool_dev);
> + cdev = cpufreq_cooling_register(data->policy);
> + if (IS_ERR(cdev)) {
> + int ret = PTR_ERR(cdev);
> dev_err(bgp->dev, "Failed to register cpu cooling device %d\n",
> ret);
> cpufreq_cpu_put(data->policy);
> @@ -277,7 +277,7 @@ int ti_thermal_unregister_cpu_cooling(struct ti_bandgap *bgp, int id)
> data = ti_bandgap_get_sensor_data(bgp, id);
>
> if (data) {
> - cpufreq_cooling_unregister(data->cool_dev);
> + cpufreq_cooling_unregister(data->policy);
> if (data->policy)
> cpufreq_cpu_put(data->policy);
> }
> diff --git a/include/linux/cpu_cooling.h b/include/linux/cpu_cooling.h
> index bae54bb7c048..89f469ee4be4 100644
> --- a/include/linux/cpu_cooling.h
> +++ b/include/linux/cpu_cooling.h
> @@ -29,9 +29,9 @@ cpufreq_cooling_register(struct cpufreq_policy *policy);
>
> /**
> * cpufreq_cooling_unregister - function to remove cpufreq cooling device.
> - * @cdev: thermal cooling device pointer.
> + * @policy: cpufreq policy
> */
> -void cpufreq_cooling_unregister(struct thermal_cooling_device *cdev);
> +void cpufreq_cooling_unregister(struct cpufreq_policy *policy);
>
> #else /* !CONFIG_CPU_THERMAL */
> static inline struct thermal_cooling_device *
> @@ -41,7 +41,7 @@ cpufreq_cooling_register(struct cpufreq_policy *policy)
> }
>
> static inline
> -void cpufreq_cooling_unregister(struct thermal_cooling_device *cdev)
> +void cpufreq_cooling_unregister(struct cpufreq_policy *policy)
> {
> return;
> }
> diff --git a/include/linux/cpufreq.h b/include/linux/cpufreq.h
> index a1467aa7f58b..ce13204df972 100644
> --- a/include/linux/cpufreq.h
> +++ b/include/linux/cpufreq.h
> @@ -144,9 +144,6 @@ struct cpufreq_policy {
>
> /* For cpufreq driver's internal use */
> void *driver_data;
> -
> - /* Pointer to the cooling device if used for thermal mitigation */
> - struct thermal_cooling_device *cdev;
> };
>
> struct cpufreq_freqs {
> --
> 2.17.1
>
^ permalink raw reply
* Re: [RFCv2 2/8] clk: imx8m-composite: Switch to determine_rate
From: Leonard Crestez @ 2019-06-28 8:56 UTC (permalink / raw)
To: Abel Vesa
Cc: Alexandre Bailon, Georgi Djakov, Stephen Boyd, Michael Turquette,
Viresh Kumar, MyungJoo Ham, Kyungmin Park, Shawn Guo,
Aisheng Dong, Fabio Estevam, Rafael J. Wysocki, Jacky Bai,
Anson Huang, Krzysztof Kozlowski, Ulf Hansson, Saravana Kannan,
kernel@pengutronix.de, dl-linux-imx, linux-pm@vger.kernel.org,
linux-clk@vger.kernel.org, linux-arm-kernel@lists.infradead.org
In-Reply-To: <20190628084521.d64d5g54zvxlsxsl@fsr-ub1664-175>
On 28.06.2019 11:45, Abel Vesa wrote:
> On 19-06-28 10:39:50, Leonard Crestez wrote:
>> This allows consumers to use min_rate max_rate.
>>
>> @@ -45,10 +45,12 @@ static unsigned long imx8m_clk_composite_divider_recalc_rate(struct clk_hw *hw,
>> divider->flags, PCG_DIV_WIDTH);
>> }
>>
>> static int imx8m_clk_composite_compute_dividers(unsigned long rate,
>> unsigned long parent_rate,
>> + unsigned long min_rate,
>> + unsigned long max_rate,
>
> You should pass on the req instead of min_rate and max_rate here.
Then I'd have to switch imx8m_clk_composite_divider_set_rate to allocate
a dummy struct clk_rate_request on the stack. It's clearer if I just
pass the minimum parameters required.
--
Regards,
Leonard
^ permalink raw reply
* Re: [RFCv2 2/8] clk: imx8m-composite: Switch to determine_rate
From: Abel Vesa @ 2019-06-28 8:45 UTC (permalink / raw)
To: Leonard Crestez
Cc: Alexandre Bailon, Georgi Djakov, Stephen Boyd, Michael Turquette,
Viresh Kumar, MyungJoo Ham, Kyungmin Park, Shawn Guo,
Aisheng Dong, Fabio Estevam, Rafael J. Wysocki, Jacky Bai,
Anson Huang, Krzysztof Kozlowski, Ulf Hansson, Saravana Kannan,
kernel@pengutronix.de, dl-linux-imx, linux-pm@vger.kernel.org,
linux-clk@vger.kernel.org, linux-arm-kernel@lists.infradead.org
In-Reply-To: <5d62b31309e6402bd9fa608730518b39af823fb3.1561707104.git.leonard.crestez@nxp.com>
On 19-06-28 10:39:50, Leonard Crestez wrote:
> This allows consumers to use min_rate max_rate.
>
> Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com>
> ---
> drivers/clk/imx/clk-composite-8m.c | 34 +++++++++++++++++++-----------
> 1 file changed, 22 insertions(+), 12 deletions(-)
>
> diff --git a/drivers/clk/imx/clk-composite-8m.c b/drivers/clk/imx/clk-composite-8m.c
> index 388bdb94f841..1be82ec08ecd 100644
> --- a/drivers/clk/imx/clk-composite-8m.c
> +++ b/drivers/clk/imx/clk-composite-8m.c
> @@ -45,10 +45,12 @@ static unsigned long imx8m_clk_composite_divider_recalc_rate(struct clk_hw *hw,
> divider->flags, PCG_DIV_WIDTH);
> }
>
> static int imx8m_clk_composite_compute_dividers(unsigned long rate,
> unsigned long parent_rate,
> + unsigned long min_rate,
> + unsigned long max_rate,
You should pass on the req instead of min_rate and max_rate here.
> int *prediv, int *postdiv)
> {
> int div1, div2;
> int error = INT_MAX;
> int ret = -EINVAL;
> @@ -56,11 +58,17 @@ static int imx8m_clk_composite_compute_dividers(unsigned long rate,
> *prediv = 1;
> *postdiv = 1;
>
> for (div1 = 1; div1 <= PCG_PREDIV_MAX; div1++) {
> for (div2 = 1; div2 <= PCG_DIV_MAX; div2++) {
> - int new_error = ((parent_rate / div1) / div2) - rate;
> + unsigned long new_rate;
> + int new_error;
> +
> + new_rate = ((parent_rate / div1) / div2);
> + if (new_rate < min_rate || new_rate > max_rate)
> + continue;
> + new_error = new_rate - rate;
>
> if (abs(new_error) < abs(error)) {
> *prediv = div1;
> *postdiv = div2;
> error = new_error;
> @@ -69,38 +77,40 @@ static int imx8m_clk_composite_compute_dividers(unsigned long rate,
> }
> }
> return ret;
> }
>
> -static long imx8m_clk_composite_divider_round_rate(struct clk_hw *hw,
> - unsigned long rate,
> - unsigned long *prate)
> +static int imx8m_clk_composite_divider_determine_rate(struct clk_hw *hw,
> + struct clk_rate_request *req)
> {
> int prediv_value;
> int div_value;
>
> - imx8m_clk_composite_compute_dividers(rate, *prate,
> - &prediv_value, &div_value);
> - rate = DIV_ROUND_UP(*prate, prediv_value);
> + imx8m_clk_composite_compute_dividers(req->rate, req->best_parent_rate,
> + req->min_rate, req->max_rate,
> + &prediv_value, &div_value);
>
> - return DIV_ROUND_UP(rate, div_value);
> + req->rate = DIV_ROUND_UP(req->best_parent_rate, prediv_value);
> + req->rate = DIV_ROUND_UP(req->rate, div_value);
>
> + return 0;
> }
>
> static int imx8m_clk_composite_divider_set_rate(struct clk_hw *hw,
> - unsigned long rate,
> - unsigned long parent_rate)
> + unsigned long rate,
> + unsigned long parent_rate)
> {
> struct clk_divider *divider = to_clk_divider(hw);
> unsigned long flags = 0;
> int prediv_value;
> int div_value;
> int ret;
> u32 val;
>
> ret = imx8m_clk_composite_compute_dividers(rate, parent_rate,
> - &prediv_value, &div_value);
> + 0, ULONG_MAX,
> + &prediv_value, &div_value);
> if (ret)
> return -EINVAL;
>
> spin_lock_irqsave(divider->lock, flags);
>
> @@ -117,11 +127,11 @@ static int imx8m_clk_composite_divider_set_rate(struct clk_hw *hw,
> return ret;
> }
>
> static const struct clk_ops imx8m_clk_composite_divider_ops = {
> .recalc_rate = imx8m_clk_composite_divider_recalc_rate,
> - .round_rate = imx8m_clk_composite_divider_round_rate,
> + .determine_rate = imx8m_clk_composite_divider_determine_rate,
> .set_rate = imx8m_clk_composite_divider_set_rate,
> };
>
> struct clk *imx8m_clk_composite_flags(const char *name,
> const char * const *parent_names,
> --
> 2.17.1
>
^ permalink raw reply
* Re: [RFC] clk: imx8mm: Add dram freq switch support
From: Leonard Crestez @ 2019-06-28 8:30 UTC (permalink / raw)
To: Stephen Boyd, Alexandre Bailon, Jacky Bai, Michael Turquette
Cc: Viresh Kumar, Anson Huang, Abel Vesa, Peng Fan, Shawn Guo,
Aisheng Dong, Fabio Estevam, Georgi Djakov, Cedric Neveux,
Silvano Di Ninno, Nitin Garg, Rafael J. Wysocki,
Krzysztof Kozlowski, linux-pm@vger.kernel.org,
linux-clk@vger.kernel.org, kernel@pengutronix.de, dl-linux-imx,
linux-arm-kernel@lists.infradead.org
In-Reply-To: <20190627211453.37FF4208CB@mail.kernel.org>
On 28.06.2019 00:15, Stephen Boyd wrote:
> Quoting Leonard Crestez (2019-05-30 00:13:51)
>> Add a wrapper clock encapsulating dram frequency switch support for
>> imx8m chips. This allows higher-level DVFS code to manipulate dram
>> frequency using standard clock framework APIs.
>>
>> Linux-side implementation is similar in principle to imx_clk_cpu or a
>> composite clock. Only some preparation is done inside the kernel, the
>> actual freq switch is performed from TF-A code which runs from an SRAM
>> area. Cores other than the one performing the switch are also made to
>> spin inside TF-A by sending each an IRQ.
>>
>> This is an early proof-of-concept which only support low/high mode on
>> imx8mm but NXP has secure-world dram freq switching implementations for
>> multiple other chips and this approach can be extended.
>>
>> Many platforms handle this kind of stuff externally but cpufreq is quite
>> insistent that actual rates are set by clk code and that new platforms
>> use cpufreq-dt.
>>
>> Let me know if there are objections to handling dram freq via clk.
>
> Can it be an interconnect driver instead? I don't see how this is a clk
> driver. It looks more like a driver that itself manages a collection of
> clks, and you've put the coordination of those clks behind the clk_ops
> interface. We don't want to have clk_ops calling clk consumer APIs in
> general, so the whole approach doesn't seem correct.
The imx8m dram clk structure is only slightly more complicated than that
for the cpu. It's not clear why mux manipulation should be pushed away
from clk and onto consumers. Isn't it desirable for clk_set_rate to
"just work"?
Implementation uses consumer APIs because the constructor takes struct
clk*. It could be modifed to take struct clk_hw*, but probably only with
larger changes to clk-imx8m.
The interrupt handling should be moved to secure world.
> Hopefully this can
> work out as some other sort of driver that is used directly from devfreq
> or interconnect core instead and then have a different consumer driver
> of devfreq or interconnect core that knows how to drive the clk tree.
Hiding dram rate setting behind a clk rate makes it much easier to
implement devfreq or interconnect, as in this series:
https://patchwork.kernel.org/project/linux-arm-kernel/list/?series=139367
I sent than RFC after you replied to this email, mostly because it's
been pending for a while.
--
Regards,
Leonard
^ permalink raw reply
* Re: [PATCH RFC 3/4] cpufreq: qcom: Update the bandwidth levels on frequency change
From: Rajendra Nayak @ 2019-06-28 8:25 UTC (permalink / raw)
To: Sibi Sankar, viresh.kumar, nm, sboyd, georgi.djakov
Cc: agross, david.brown, robh+dt, mark.rutland, rjw, linux-arm-msm,
devicetree, linux-kernel, linux-pm, saravanak
In-Reply-To: <20190627133424.4980-4-sibis@codeaurora.org>
On 6/27/2019 7:04 PM, Sibi Sankar wrote:
> Add support to parse and update optional OPP tables attached to the
> cpu nodes when the OPP bandwidth values are populated to enable
> scaling of DDR/L3 bandwidth levels with frequency change.
>
> Signed-off-by: Sibi Sankar <sibis@codeaurora.org>
> ---
[]...
>
> @@ -79,13 +108,29 @@ static int qcom_cpufreq_hw_read_lut(struct device *cpu_dev,
> {
> u32 data, src, lval, i, core_count, prev_cc = 0, prev_freq = 0, freq;
> u32 volt;
> + u64 rate;
> unsigned int max_cores = cpumask_weight(policy->cpus);
> struct cpufreq_frequency_table *table;
> + struct device_node *opp_table_np, *np;
> + int ret;
>
> table = kcalloc(LUT_MAX_ENTRIES + 1, sizeof(*table), GFP_KERNEL);
> if (!table)
> return -ENOMEM;
>
> + ret = dev_pm_opp_of_add_table(cpu_dev);
> + if (!ret) {
> + /* Disable all opps and cross-validate against LUT */
> + opp_table_np = dev_pm_opp_of_get_opp_desc_node(cpu_dev);
> + for_each_available_child_of_node(opp_table_np, np) {
> + ret = of_property_read_u64(np, "opp-hz", &rate);
> + dev_pm_opp_disable(cpu_dev, rate);
> + }
> + of_node_put(opp_table_np);
> + } else {
> + dev_err(cpu_dev, "Couldn't add OPP table from dt\n");
The changelog seems to suggest specifying OPP tables in DT is optional,
but here we seem to error out if the tables are missing.
> + }
> +
> for (i = 0; i < LUT_MAX_ENTRIES; i++) {
> data = readl_relaxed(base + REG_FREQ_LUT +
> i * LUT_ROW_SIZE);
> @@ -104,7 +149,7 @@ static int qcom_cpufreq_hw_read_lut(struct device *cpu_dev,
>
> if (freq != prev_freq && core_count == max_cores) {
> table[i].frequency = freq;
> - dev_pm_opp_add(cpu_dev, freq * 1000, volt);
> + qcom_cpufreq_update_opp(cpu_dev, freq, volt);
> dev_dbg(cpu_dev, "index=%d freq=%d, core_count %d\n", i,
> freq, core_count);
> } else {
> @@ -125,7 +170,8 @@ static int qcom_cpufreq_hw_read_lut(struct device *cpu_dev,
> if (prev_cc != max_cores) {
> prev->frequency = prev_freq;
> prev->flags = CPUFREQ_BOOST_FREQ;
> - dev_pm_opp_add(cpu_dev, prev_freq * 1000, volt);
> + qcom_cpufreq_update_opp(cpu_dev, prev_freq,
> + volt);
> }
>
> break;
> @@ -168,6 +214,7 @@ static void qcom_get_related_cpus(int index, struct cpumask *m)
> static int qcom_cpufreq_hw_cpu_init(struct cpufreq_policy *policy)
> {
> struct device *dev = &global_pdev->dev;
> + struct opp_table *opp_table = NULL;
> struct of_phandle_args args;
> struct device_node *cpu_np;
> struct device *cpu_dev;
> @@ -202,6 +249,8 @@ static int qcom_cpufreq_hw_cpu_init(struct cpufreq_policy *policy)
> if (!base)
> return -ENOMEM;
>
> + opp_table = dev_pm_opp_set_paths(cpu_dev);
> +
> /* HW should be in enabled state to proceed */
> if (!(readl_relaxed(base + REG_ENABLE) & 0x1)) {
> dev_err(dev, "Domain-%d cpufreq hardware not enabled\n", index);
> @@ -237,6 +286,8 @@ static int qcom_cpufreq_hw_cpu_init(struct cpufreq_policy *policy)
>
> return 0;
> error:
> + if (opp_table)
> + dev_pm_opp_put_paths(opp_table);
> devm_iounmap(dev, base);
> return ret;
> }
> @@ -275,6 +326,8 @@ static struct cpufreq_driver cpufreq_qcom_hw_driver = {
>
> static int qcom_cpufreq_hw_driver_probe(struct platform_device *pdev)
> {
> + struct opp_table *opp_table = NULL;
> + struct device *cpu_dev;
> struct clk *clk;
> int ret;
>
> @@ -294,6 +347,26 @@ static int qcom_cpufreq_hw_driver_probe(struct platform_device *pdev)
>
> global_pdev = pdev;
>
> + /* Check for optional interconnect paths on CPU0 */
> + cpu_dev = get_cpu_device(0);
> + if (!cpu_dev) {
> + dev_err(&pdev->dev, "failed to get cpu0 device\n");
> + return -ENODEV;
> + }
> +
> + opp_table = dev_pm_opp_set_paths(cpu_dev);
> + if (IS_ERR(opp_table)) {
> + ret = PTR_ERR(opp_table);
> + if (ret == -EPROBE_DEFER) {
> + dev_dbg(&pdev->dev, "defer icc set paths: %d\n", ret);
> + return ret;
> + }
> + dev_err(&pdev->dev, "set paths failed ddr/l3 scaling off: %d\n",
> + ret);
Here again, the interconnect paths don't seem to be optional as the comment
above suggests.
> + } else {
> + dev_pm_opp_put_paths(opp_table);
> + }
> +
> ret = cpufreq_register_driver(&cpufreq_qcom_hw_driver);
> if (ret)
> dev_err(&pdev->dev, "CPUFreq HW driver failed to register\n");
>
--
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member
of Code Aurora Forum, hosted by The Linux Foundation
^ permalink raw reply
* Re: [PATCH 2/2] powercap/rapl: Add Ice Lake NNPI support to RAPL driver
From: Bhardwaj, Rajneesh @ 2019-06-28 8:21 UTC (permalink / raw)
To: Rafael J. Wysocki; +Cc: linux-kernel, len.brown, rafael.j.wysocki, linux-pm
In-Reply-To: <2359736.lHFMKXy7DL@kreacher>
On 28-Jun-19 3:32 AM, Rafael J. Wysocki wrote:
> On Friday, June 14, 2019 10:05:23 AM CEST Rajneesh Bhardwaj wrote:
>> Enables support for ICL-NNPI, which is a neural network processor for deep
>> learning inference. From RAPL point of view it is same as Ice Lake Mobile
>> processor.
>>
>> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net>
>> Cc: linux-pm@vger.kernel.org
>> Link: https://lkml.org/lkml/2019/6/5/1034
>> Signed-off-by: Rajneesh Bhardwaj <rajneesh.bhardwaj@linux.intel.com>
>> ---
>> drivers/powercap/intel_rapl.c | 1 +
>> 1 file changed, 1 insertion(+)
>>
>> diff --git a/drivers/powercap/intel_rapl.c b/drivers/powercap/intel_rapl.c
>> index 4347f15165f8..431c8c8bdf07 100644
>> --- a/drivers/powercap/intel_rapl.c
>> +++ b/drivers/powercap/intel_rapl.c
>> @@ -1157,6 +1157,7 @@ static const struct x86_cpu_id rapl_ids[] __initconst = {
>> INTEL_CPU_FAM6(KABYLAKE_DESKTOP, rapl_defaults_core),
>> INTEL_CPU_FAM6(CANNONLAKE_MOBILE, rapl_defaults_core),
>> INTEL_CPU_FAM6(ICELAKE_MOBILE, rapl_defaults_core),
>> + INTEL_CPU_FAM6(ICELAKE_NNPI, rapl_defaults_core),
>>
>> INTEL_CPU_FAM6(ATOM_SILVERMONT, rapl_defaults_byt),
>> INTEL_CPU_FAM6(ATOM_AIRMONT, rapl_defaults_cht),
>>
> It is in my queue, but I get build errors when I try to apply it.
>
> I guess the definition of ICELAKE_NNPI is not there in the Linus' tree yet.
Yes, though
https://www.spinics.net/lists/linux-tip-commits/msg48978.html is
accepted by Thomas already.
>
>
>
^ permalink raw reply
* Re: [PATCH RFC 1/4] OPP: Add and export helper to update voltage
From: Rajendra Nayak @ 2019-06-28 8:20 UTC (permalink / raw)
To: Sibi Sankar, viresh.kumar, nm, sboyd, georgi.djakov
Cc: agross, david.brown, robh+dt, mark.rutland, rjw, linux-arm-msm,
devicetree, linux-kernel, linux-pm, saravanak
In-Reply-To: <20190627133424.4980-2-sibis@codeaurora.org>
On 6/27/2019 7:04 PM, Sibi Sankar wrote:
> Add and export 'dev_pm_opp_update_voltage' to find and update voltage
> of an opp for a given frequency. This will be useful to update the opps
> with voltages read back from firmware.
>
> Signed-off-by: Sibi Sankar <sibis@codeaurora.org>
> ---
> drivers/opp/core.c | 52 ++++++++++++++++++++++++++++++++++++++++++
> include/linux/pm_opp.h | 10 ++++++++
> 2 files changed, 62 insertions(+)
>
> diff --git a/drivers/opp/core.c b/drivers/opp/core.c
> index 68551d6366e6b..c85c04dc2c7de 100644
> --- a/drivers/opp/core.c
> +++ b/drivers/opp/core.c
> @@ -2197,6 +2197,58 @@ int dev_pm_opp_disable(struct device *dev, unsigned long freq)
> }
> EXPORT_SYMBOL_GPL(dev_pm_opp_disable);
>
> +/**
> + * dev_pm_opp_update_voltage() - Find and update voltage
> + * @dev: device for which we do this operation
> + * @freq: OPP frequency to update voltage
> + * @u_volt: voltage requested for this opp
> + *
> + * Find and update voltage of a disabled opp corresponding to the given
> + * frequency. This is useful only for devices with single power supply.
> + *
> + * Return: 0 if no modification was done OR modification was
> + * successful or a negative error value.
> + */
> +int dev_pm_opp_update_voltage(struct device *dev, unsigned long freq,
> + unsigned long u_volt)
> +{
> + struct dev_pm_opp *opp = ERR_PTR(-ENODEV);
> + struct opp_table *opp_table;
> + unsigned long tol;
> + int ret = 0;
> +
> + opp = dev_pm_opp_find_freq_exact(dev, freq, false);
> + if (IS_ERR(opp))
> + return PTR_ERR(opp);
> +
> + /* Find the opp_table */
> + opp_table = _find_opp_table(dev);
maybe you should look for the opp_table first and then the
opp? otherwise this check seems pretty redundant given the
above call would have failed if the opp table did not exist.
> + if (IS_ERR(opp_table)) {
> + ret = PTR_ERR(opp_table);
> + dev_err(dev, "%s: OPP table not found (%d)\n", __func__, ret);
> + goto put_opp;
> + }
> +
> + mutex_lock(&opp_table->lock);
> +
> + /* update only if the opp is disabled */
> + if (opp->available)
> + goto unlock;
should this throw a warning/error to indicate someone tried to
update this at a wrong time? and perhaps return an error instead of
success.
> +
> + tol = u_volt * opp_table->voltage_tolerance_v1 / 100;
> + opp->supplies[0].u_volt_min = u_volt - tol;
> + opp->supplies[0].u_volt = u_volt;
> + opp->supplies[0].u_volt_min = u_volt + tol;
> +
> +unlock:
> + mutex_unlock(&opp_table->lock);
> + dev_pm_opp_put_opp_table(opp_table);
> +put_opp:
> + dev_pm_opp_put(opp);
> + return ret;
> +}
> +EXPORT_SYMBOL_GPL(dev_pm_opp_update_voltage);
> +
> /**
> * dev_pm_opp_register_notifier() - Register OPP notifier for the device
> * @dev: Device for which notifier needs to be registered
> diff --git a/include/linux/pm_opp.h b/include/linux/pm_opp.h
> index 87fa09d93d8c2..a17c462974851 100644
> --- a/include/linux/pm_opp.h
> +++ b/include/linux/pm_opp.h
> @@ -130,6 +130,9 @@ int dev_pm_opp_enable(struct device *dev, unsigned long freq);
>
> int dev_pm_opp_disable(struct device *dev, unsigned long freq);
>
> +int dev_pm_opp_update_voltage(struct device *dev, unsigned long freq,
> + unsigned long u_volt);
> +
> int dev_pm_opp_register_notifier(struct device *dev, struct notifier_block *nb);
> int dev_pm_opp_unregister_notifier(struct device *dev, struct notifier_block *nb);
>
> @@ -261,6 +264,13 @@ static inline int dev_pm_opp_disable(struct device *dev, unsigned long freq)
> return 0;
> }
>
> +static inline int dev_pm_opp_update_voltage(struct device *dev,
> + unsigned long freq,
> + unsigned long u_volt)
> +{
> + return 0;
> +}
> +
> static inline int dev_pm_opp_register_notifier(struct device *dev, struct notifier_block *nb)
> {
> return -ENOTSUPP;
>
--
QUALCOMM INDIA, on behalf of Qualcomm Innovation Center, Inc. is a member
of Code Aurora Forum, hosted by The Linux Foundation
^ permalink raw reply
* [RFCv2 7/8] arm64: dts: imx8mm: Add interconnect node
From: Leonard Crestez @ 2019-06-28 7:39 UTC (permalink / raw)
To: Alexandre Bailon, Georgi Djakov, Stephen Boyd, Michael Turquette,
Viresh Kumar
Cc: MyungJoo Ham, Kyungmin Park, Shawn Guo, Dong Aisheng,
Fabio Estevam, Rafael J. Wysocki, Jacky Bai, Anson Huang,
Abel Vesa, Krzysztof Kozlowski, Ulf Hansson, Saravana Kannan,
kernel, linux-imx, linux-pm, linux-clk, linux-arm-kernel
In-Reply-To: <cover.1561707104.git.leonard.crestez@nxp.com>
Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com>
---
arch/arm64/boot/dts/freescale/imx8mm.dtsi | 11 +++++++++++
1 file changed, 11 insertions(+)
diff --git a/arch/arm64/boot/dts/freescale/imx8mm.dtsi b/arch/arm64/boot/dts/freescale/imx8mm.dtsi
index 5da905c257ad..3b4b112814f7 100644
--- a/arch/arm64/boot/dts/freescale/imx8mm.dtsi
+++ b/arch/arm64/boot/dts/freescale/imx8mm.dtsi
@@ -6,10 +6,11 @@
#include <dt-bindings/clock/imx8mm-clock.h>
#include <dt-bindings/gpio/gpio.h>
#include <dt-bindings/input/input.h>
#include <dt-bindings/interrupt-controller/arm-gic.h>
#include <dt-bindings/thermal/thermal.h>
+#include <dt-bindings/interconnect/imx8mm.h>
#include "imx8mm-pinfunc.h"
/ {
compatible = "fsl,imx8mm";
@@ -179,10 +180,20 @@
interrupts = <GIC_PPI 7
(GIC_CPU_MASK_SIMPLE(6) | IRQ_TYPE_LEVEL_HIGH)>;
interrupt-affinity = <&A53_0>, <&A53_1>, <&A53_2>, <&A53_3>;
};
+ icc: interconnect {
+ compatible = "fsl,imx8mm-interconnect";
+ #interconnect-cells = <1>;
+ clocks = <&clk IMX8MM_CLK_DRAM>,
+ <&clk IMX8MM_CLK_NOC>,
+ <&clk IMX8MM_CLK_AHB>,
+ <&clk IMX8MM_CLK_MAIN_AXI>;
+ clock-names = "dram", "noc", "ahb", "axi";
+ };
+
timer {
compatible = "arm,armv8-timer";
interrupts = <GIC_PPI 13 (GIC_CPU_MASK_SIMPLE(6) | IRQ_TYPE_LEVEL_LOW)>, /* Physical Secure */
<GIC_PPI 14 (GIC_CPU_MASK_SIMPLE(6) | IRQ_TYPE_LEVEL_LOW)>, /* Physical Non-Secure */
<GIC_PPI 11 (GIC_CPU_MASK_SIMPLE(6) | IRQ_TYPE_LEVEL_LOW)>, /* Virtual */
--
2.17.1
^ permalink raw reply related
* [RFCv2 8/8] arm64: dts: imx8mm: Add devfreq-imx nodes
From: Leonard Crestez @ 2019-06-28 7:39 UTC (permalink / raw)
To: Alexandre Bailon, Georgi Djakov, Stephen Boyd, Michael Turquette,
Viresh Kumar
Cc: MyungJoo Ham, Kyungmin Park, Shawn Guo, Dong Aisheng,
Fabio Estevam, Rafael J. Wysocki, Jacky Bai, Anson Huang,
Abel Vesa, Krzysztof Kozlowski, Ulf Hansson, Saravana Kannan,
kernel, linux-imx, linux-pm, linux-clk, linux-arm-kernel
In-Reply-To: <cover.1561707104.git.leonard.crestez@nxp.com>
The imx8mm has multiple buses which can be scaled with some degree of
independence. Expose them as devfreq devices for userspace scaling.
It shouldn't be possible to get the system in a non-working state this
way. It is primarily aimed at testing and fine performance tuning.
Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com>
---
arch/arm64/boot/dts/freescale/imx8mm.dtsi | 54 +++++++++++++++++++++++
1 file changed, 54 insertions(+)
diff --git a/arch/arm64/boot/dts/freescale/imx8mm.dtsi b/arch/arm64/boot/dts/freescale/imx8mm.dtsi
index 3b4b112814f7..aa9ed418652d 100644
--- a/arch/arm64/boot/dts/freescale/imx8mm.dtsi
+++ b/arch/arm64/boot/dts/freescale/imx8mm.dtsi
@@ -121,10 +121,32 @@
opp-supported-hw = <0x8>, <0x7>;
clock-latency-ns = <150000>;
};
};
+ ddrc_opp_table: ddrc-opp-table {
+ compatible = "operating-points-v2";
+
+ opp-25M {
+ opp-hz = /bits/ 64 <25000000>;
+ };
+ opp-750M {
+ opp-hz = /bits/ 64 <750000000>;
+ };
+ };
+
+ noc_opp_table: noc-opp-table {
+ compatible = "operating-points-v2";
+
+ opp-150M {
+ opp-hz = /bits/ 64 <150000000>;
+ };
+ opp-750M {
+ opp-hz = /bits/ 64 <750000000>;
+ };
+ };
+
memory@40000000 {
device_type = "memory";
reg = <0x0 0x40000000 0 0x80000000>;
};
@@ -748,10 +770,35 @@
status = "disabled";
};
};
+ pl301_main: nic@32000000 {
+ compatible = "fsl,imx8mm-nic";
+ reg = <0x32000000 0x100000>;
+ clocks = <&clk IMX8MM_CLK_MAIN_AXI>;
+ };
+
+ pl301_wakeup: nic@32100000 {
+ compatible = "fsl,imx8mm-nic";
+ reg = <0x32100000 0x100000>;
+ clocks = <&clk IMX8MM_CLK_AHB>;
+ };
+
+ pl301_enet: nic@32400000 {
+ compatible = "fsl,imx8mm-nic";
+ reg = <0x32400000 0x100000>;
+ clocks = <&clk IMX8MM_CLK_ENET_AXI>;
+ };
+
+ noc: noc@32700000 {
+ compatible = "fsl,imx8mm-noc";
+ reg = <0x32700000 0x100000>;
+ clocks = <&clk IMX8MM_CLK_NOC>;
+ operating-points-v2 = <&noc_opp_table>;
+ };
+
aips4: bus@32c00000 {
compatible = "fsl,aips-bus", "simple-bus";
#address-cells = <1>;
#size-cells = <1>;
ranges = <0x32c00000 0x32c00000 0x400000>;
@@ -835,7 +882,14 @@
<0x38880000 0xc0000>; /* GICR (RD_base + SGI_base) */
#interrupt-cells = <3>;
interrupt-controller;
interrupts = <GIC_PPI 9 IRQ_TYPE_LEVEL_HIGH>;
};
+
+ ddrc: dram-controller@3d400000 {
+ compatible = "fsl,imx8mm-ddrc";
+ reg = <0x3d400000 0x400000>;
+ clocks = <&clk IMX8MM_CLK_DRAM>;
+ operating-points-v2 = <&ddrc_opp_table>;
+ };
};
};
--
2.17.1
^ permalink raw reply related
* [RFCv2 4/8] interconnect: Add generic driver for imx
From: Leonard Crestez @ 2019-06-28 7:39 UTC (permalink / raw)
To: Alexandre Bailon, Georgi Djakov, Stephen Boyd, Michael Turquette,
Viresh Kumar
Cc: MyungJoo Ham, Kyungmin Park, Shawn Guo, Dong Aisheng,
Fabio Estevam, Rafael J. Wysocki, Jacky Bai, Anson Huang,
Abel Vesa, Krzysztof Kozlowski, Ulf Hansson, Saravana Kannan,
kernel, linux-imx, linux-pm, linux-clk, linux-arm-kernel
In-Reply-To: <cover.1561707104.git.leonard.crestez@nxp.com>
From: Alexandre Bailon <abailon@baylibre.com>
This adds support for i.MX SoC to interconnect framework.
This is based on busfreq from NXP vendor tree tree. The implementation
is sufficiently generic to be applies to any imx SOC and perhaps others.
Using the interconnect framework driver requests for bandwidth are used
to determine a global "platform opp" which is then and then scale the
frequency. These "platform opps" are similar to the "low/audio/high"
busfreq levels from NXP tree.
Busfreq platform drivers just have to register interconnect nodes and
OPPs.
Signed-off-by: Alexandre Bailon <abailon@baylibre.com>
Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com>
Changes by Leonard:
* Make it probe: add xlate and provider->dev
* Clarify commit message
* Use unsigned long for rate
Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com>
---
drivers/interconnect/Kconfig | 1 +
drivers/interconnect/Makefile | 1 +
drivers/interconnect/imx/Kconfig | 13 +
drivers/interconnect/imx/Makefile | 1 +
drivers/interconnect/imx/busfreq.c | 628 +++++++++++++++++++++++++++++
drivers/interconnect/imx/busfreq.h | 123 ++++++
6 files changed, 767 insertions(+)
create mode 100644 drivers/interconnect/imx/Kconfig
create mode 100644 drivers/interconnect/imx/Makefile
create mode 100644 drivers/interconnect/imx/busfreq.c
create mode 100644 drivers/interconnect/imx/busfreq.h
diff --git a/drivers/interconnect/Kconfig b/drivers/interconnect/Kconfig
index bfa4ca3ab7a9..e61802230f90 100644
--- a/drivers/interconnect/Kconfig
+++ b/drivers/interconnect/Kconfig
@@ -10,7 +10,8 @@ menuconfig INTERCONNECT
If unsure, say no.
if INTERCONNECT
source "drivers/interconnect/qcom/Kconfig"
+source "drivers/interconnect/imx/Kconfig"
endif
diff --git a/drivers/interconnect/Makefile b/drivers/interconnect/Makefile
index 28f2ab0824d5..20a13b7eb37f 100644
--- a/drivers/interconnect/Makefile
+++ b/drivers/interconnect/Makefile
@@ -2,5 +2,6 @@
icc-core-objs := core.o
obj-$(CONFIG_INTERCONNECT) += icc-core.o
obj-$(CONFIG_INTERCONNECT_QCOM) += qcom/
+obj-$(CONFIG_INTERCONNECT_IMX) += imx/
diff --git a/drivers/interconnect/imx/Kconfig b/drivers/interconnect/imx/Kconfig
new file mode 100644
index 000000000000..afd7b71bbd82
--- /dev/null
+++ b/drivers/interconnect/imx/Kconfig
@@ -0,0 +1,13 @@
+config INTERCONNECT_IMX
+ bool "i.MX interconnect drivers"
+ depends on ARCH_MXC || ARCH_MXC_ARM64 || COMPILE_TEST
+ help
+ Support for i.MX interconnect hardware.
+
+config BUSFREQ
+ bool "busfreq interconnect driver"
+ depends on INTERCONNECT_IMX
+ help
+ A generic interconnect driver that could be used for any i.MX.
+ This provides a way to register master and slave and some opp
+ to use when one or more master are in use.
diff --git a/drivers/interconnect/imx/Makefile b/drivers/interconnect/imx/Makefile
new file mode 100644
index 000000000000..fea647183815
--- /dev/null
+++ b/drivers/interconnect/imx/Makefile
@@ -0,0 +1 @@
+obj-$(CONFIG_BUSFREQ) += busfreq.o
diff --git a/drivers/interconnect/imx/busfreq.c b/drivers/interconnect/imx/busfreq.c
new file mode 100644
index 000000000000..19578728e1b3
--- /dev/null
+++ b/drivers/interconnect/imx/busfreq.c
@@ -0,0 +1,628 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Interconnect framework driver for i.MX SoC
+ *
+ * Copyright (c) 2019, BayLibre
+ * Author: Alexandre Bailon <abailon@baylibre.com>
+ */
+
+#include <linux/clk.h>
+#include <linux/device.h>
+#include <linux/interconnect-provider.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/reboot.h>
+#include <linux/suspend.h>
+#include <linux/of.h>
+
+#include "busfreq.h"
+
+/*
+ * struct busfreq_opp_node - Describe the minimum bandwidth required by a node
+ * to enable the opp
+ * @icc_node: icc_node to test
+ * @min_avg_bw: minimum average bandwidth in kbps required to enable opp
+ * @min_peak_bw: minimum peak bandwidth in kbps required to enable opp
+ */
+struct busfreq_opp_node {
+ struct icc_node *icc_node;
+ u32 min_avg_bw;
+ u32 min_peak_bw;
+};
+
+/*
+ * struct busfreq_opp - Describe an opp
+ * This is used to configure multiple clock at once with the respect of
+ * hardware and performance requirements.
+ * @clks_count: number of clocks to configure
+ * @clks: array of clock
+ * @rates: array of rate, to apply for each clock when the opp is enabled
+ * @perf_level: Used to select the opp that would allow the lowest power
+ * consumption when two or more opp satisfies the performances
+ * requirements
+ * @node: entry in list of opp
+ * @nodes_count: number of opp node
+ * @nodes: array of opp node, to check node by node if opp satisfies the
+ * the required performances
+ */
+struct busfreq_opp {
+ int clks_count;
+ struct clk **clks;
+ unsigned long *rates;
+ u32 perf_level;
+
+ struct list_head node;
+
+ int nodes_count;
+ struct busfreq_opp_node *nodes;
+};
+
+/*
+ * struct busfreq_icc_desc - Hold data required to control the interconnects
+ * @dev: device pointer for the overall interconnect
+ * @opps: list of opp
+ * @default_opp: the opp opp to use when the system is in special states
+ * (boot, suspend, resume, shutdown)
+ * @current_opp: the opp currently in use
+ * @opp_locked: prevent opp to change while this is set
+ * @pm_notifier: used to set the default opp before suspend and
+ * and select the best one after resume
+ * @pm_notifier: used to set the default opp before to reboot
+ */
+struct busfreq_icc_desc {
+ struct device *dev;
+
+ struct list_head opps;
+ struct busfreq_opp *default_opp;
+ struct busfreq_opp *current_opp;
+ bool opp_locked;
+
+ struct notifier_block pm_notifier;
+ struct notifier_block reboot_notifier;
+
+ struct mutex mutex;
+};
+
+static int busfreq_icc_aggregate(struct icc_node *node, u32 avg_bw,
+ u32 peak_bw, u32 *agg_avg, u32 *agg_peak)
+{
+ *agg_avg += avg_bw;
+ *agg_peak = max(*agg_peak, peak_bw);
+
+ return 0;
+}
+
+static struct icc_node* busfreq_icc_xlate(struct of_phandle_args *spec, void *data)
+{
+ struct busfreq_icc_desc *desc = data;
+ struct icc_provider *provider = dev_get_drvdata(desc->dev);
+ unsigned int id = spec->args[0];
+ struct icc_node *node;
+
+ list_for_each_entry (node, &provider->nodes, node_list)
+ if (node->id == id)
+ return node;
+
+ return ERR_PTR(-EINVAL);
+}
+
+static int busfreq_doset_opp(struct busfreq_opp *opp)
+{
+ int i;
+ int ret;
+
+ for (i = 0; i < opp->clks_count; i++) {
+ ret = clk_set_min_rate(opp->clks[i], opp->rates[i]);
+ if (ret)
+ goto err;
+ }
+
+ return 0;
+err:
+ for (; i >= 0; i--)
+ clk_set_min_rate(opp->clks[i], 0);
+
+ return ret;
+}
+
+static void busfreq_unset_opp(struct busfreq_opp *opp)
+{
+ int i;
+
+ for (i = 0; i < opp->clks_count; i++)
+ clk_set_min_rate(opp->clks[i], 0);
+}
+
+static int busfreq_set_opp_no_lock(struct busfreq_icc_desc *desc,
+ struct busfreq_opp *requested_opp)
+{
+ int ret;
+ int i;
+
+ if (!requested_opp && !desc->default_opp)
+ return -EINVAL;
+
+ if (!requested_opp)
+ requested_opp = desc->default_opp;
+
+ if (desc->current_opp == requested_opp)
+ return 0;
+
+ if (desc->opp_locked)
+ return -EBUSY;
+
+ /*
+ * Each OPP fetches clks separately and min_rate is aggregated by ccf
+ * so we need to "unset" the min_rate explicitly afterwards.
+ */
+ ret = busfreq_doset_opp(requested_opp);
+ if (ret)
+ return ret;
+
+ if (desc->current_opp)
+ busfreq_unset_opp(desc->current_opp);
+ desc->current_opp = requested_opp;
+
+ for (i = 0; i < requested_opp->clks_count; i++) {
+ /* clk must implement determine_rate to deal with min_rate */
+ ret = clk_set_rate(requested_opp->clks[i], 0);
+ if (ret)
+ dev_err(desc->dev, "failed to request min rate\n");
+ }
+
+ /* Read back for debugging: */
+ for (i = 0; i < requested_opp->clks_count; i++) {
+ unsigned long new_rate, round_zero_rate;
+ new_rate = clk_get_rate(requested_opp->clks[i]);
+ round_zero_rate = clk_round_rate(requested_opp->clks[i], 0);
+ dev_dbg(desc->dev, "clk[%d] rate %lu round0 %lu set_min %lu\n",
+ i, new_rate, round_zero_rate,
+ requested_opp->rates[i]);
+ }
+
+ return 0;
+}
+
+static int busfreq_set_opp(struct busfreq_icc_desc *desc,
+ struct busfreq_opp *requested_opp)
+{
+ int ret;
+
+ mutex_lock(&desc->mutex);
+ ret = busfreq_set_opp_no_lock(desc, requested_opp);
+ mutex_unlock(&desc->mutex);
+
+ return ret;
+}
+
+static int busfreq_opp_bw_gt(struct busfreq_opp_node *opp_node,
+ u32 avg_bw, u32 peak_bw)
+{
+ if (!opp_node)
+ return 0;
+ if (opp_node->min_avg_bw == BUSFREQ_UNDEFINED_BW) {
+ if (avg_bw)
+ return 2;
+ else
+ return 1;
+ }
+ if (opp_node->min_peak_bw == BUSFREQ_UNDEFINED_BW) {
+ if (peak_bw)
+ return 2;
+ else
+ return 1;
+ }
+ if (avg_bw >= opp_node->min_avg_bw)
+ return 1;
+ if (peak_bw >= opp_node->min_peak_bw)
+ return 1;
+ return 0;
+}
+
+static struct busfreq_opp *busfreq_cmp_bw_opp(struct busfreq_icc_desc *desc,
+ struct busfreq_opp *opp1,
+ struct busfreq_opp *opp2)
+{
+ int i;
+ int opp1_valid;
+ int opp2_valid;
+ int opp1_count = 0;
+ int opp2_count = 0;
+
+ if (!opp1 && !opp2)
+ return desc->current_opp;
+
+ if (!opp1)
+ return opp2;
+
+ if (!opp2)
+ return opp1;
+
+ if (opp1 == opp2)
+ return opp1;
+
+ for (i = 0; i < opp1->nodes_count; i++) {
+ struct busfreq_opp_node *opp_node1, *opp_node2;
+ struct icc_node *icc_node;
+ u32 avg_bw;
+ u32 peak_bw;
+
+ opp_node1 = &opp1->nodes[i];
+ opp_node2 = &opp2->nodes[i];
+ icc_node = opp_node1->icc_node;
+ avg_bw = icc_node->avg_bw;
+ peak_bw = icc_node->peak_bw;
+
+ opp1_valid = busfreq_opp_bw_gt(opp_node1, avg_bw, peak_bw);
+ opp2_valid = busfreq_opp_bw_gt(opp_node2, avg_bw, peak_bw);
+
+ if (opp1_valid == opp2_valid && opp1_valid == 1) {
+ if (opp_node1->min_avg_bw > opp_node2->min_avg_bw &&
+ opp_node1->min_avg_bw != BUSFREQ_UNDEFINED_BW)
+ opp1_valid++;
+ if (opp_node1->min_avg_bw < opp_node2->min_avg_bw &&
+ opp_node2->min_avg_bw != BUSFREQ_UNDEFINED_BW)
+ opp2_valid++;
+ }
+
+ opp1_count += opp1_valid;
+ opp2_count += opp2_valid;
+ }
+
+ if (opp1_count > opp2_count)
+ return opp1;
+ if (opp1_count < opp2_count)
+ return opp2;
+ return opp1->perf_level >= opp2->perf_level ? opp2 : opp1;
+}
+
+static int busfreq_set_best_opp(struct busfreq_icc_desc *desc)
+{
+ struct busfreq_opp *opp, *best_opp = desc->current_opp;
+
+ list_for_each_entry(opp, &desc->opps, node)
+ best_opp = busfreq_cmp_bw_opp(desc, opp, best_opp);
+ return busfreq_set_opp(desc, best_opp);
+}
+
+static int busfreq_set_locked_opp(struct busfreq_icc_desc *desc,
+ struct busfreq_opp *requested_opp)
+{
+ int ret;
+
+ mutex_lock(&desc->mutex);
+ ret = busfreq_set_opp_no_lock(desc, requested_opp);
+ if (ret)
+ goto err;
+ desc->opp_locked = true;
+err:
+ mutex_unlock(&desc->mutex);
+
+ return ret;
+}
+
+static int busfreq_unlock_opp(struct busfreq_icc_desc *desc)
+{
+ mutex_lock(&desc->mutex);
+ desc->opp_locked = false;
+ mutex_unlock(&desc->mutex);
+
+ return busfreq_set_best_opp(desc);
+}
+
+static int busfreq_icc_set(struct icc_node *src, struct icc_node *dst)
+{
+ struct busfreq_icc_desc *desc = src->provider->data;
+
+ if (!dst->num_links)
+ return busfreq_set_best_opp(desc);
+
+ return 0;
+}
+
+static int busfreq_pm_notify(struct notifier_block *nb, unsigned long event,
+ void *ptr)
+{
+ struct busfreq_icc_desc *desc;
+
+ desc = container_of(nb, struct busfreq_icc_desc, pm_notifier);
+ if (event == PM_SUSPEND_PREPARE)
+ busfreq_set_locked_opp(desc, desc->default_opp);
+ else if (event == PM_POST_SUSPEND)
+ busfreq_unlock_opp(desc);
+
+ return NOTIFY_OK;
+}
+
+static int busfreq_reboot_notify(struct notifier_block *nb, unsigned long event,
+ void *ptr)
+{
+ struct busfreq_icc_desc *desc;
+
+ desc = container_of(nb, struct busfreq_icc_desc, reboot_notifier);
+ busfreq_set_locked_opp(desc, desc->default_opp);
+
+ return NOTIFY_OK;
+}
+
+static struct icc_node *busfreq_icc_node_add(struct icc_provider *provider,
+ int id, const char *name)
+{
+ struct busfreq_icc_desc *desc = provider->data;
+ struct device *dev = desc->dev;
+ struct icc_node *icc_node;
+
+ icc_node = icc_node_create(id);
+ if (IS_ERR(icc_node)) {
+ dev_err(dev, "Failed to create node %d\n", id);
+ return icc_node;
+ }
+
+ if (icc_node->data)
+ return icc_node;
+
+ icc_node->name = name;
+ icc_node->data = &icc_node;
+ icc_node_add(icc_node, provider);
+
+ return icc_node;
+}
+
+static struct icc_node *busfreq_icc_node_get(struct icc_provider *provider,
+ int id)
+{
+ return busfreq_icc_node_add(provider, id, NULL);
+}
+
+static void busfreq_unregister_nodes(struct icc_provider *provider)
+{
+ struct icc_node *icc_node, *tmp;
+
+ list_for_each_entry_safe(icc_node, tmp, &provider->nodes, node_list)
+ icc_node_destroy(icc_node->id);
+}
+
+static int busfreq_register_nodes(struct icc_provider *provider,
+ struct busfreq_icc_node *busfreq_nodes,
+ int count)
+{
+ int ret;
+ int i;
+
+ for (i = 0; i < count; i++) {
+ struct icc_node *icc_node;
+ size_t j;
+
+ icc_node = busfreq_icc_node_add(provider,
+ busfreq_nodes[i].id,
+ busfreq_nodes[i].name);
+ if (IS_ERR(icc_node)) {
+ ret = PTR_ERR(icc_node);
+ goto err;
+ }
+
+ for (j = 0; j < busfreq_nodes[i].num_links; j++)
+ icc_link_create(icc_node, busfreq_nodes[i].links[j]);
+ }
+
+ return 0;
+
+err:
+ busfreq_unregister_nodes(provider);
+
+ return ret;
+}
+
+static struct busfreq_opp *busfreq_opp_alloc(struct icc_provider *provider,
+ int count)
+{
+ struct busfreq_icc_desc *desc = provider->data;
+ struct device *dev = desc->dev;
+ struct busfreq_opp *opp;
+ struct icc_node *icc_node;
+
+ opp = devm_kzalloc(dev, sizeof(*opp), GFP_KERNEL);
+ if (!opp)
+ return ERR_PTR(-ENOMEM);
+
+ opp->clks_count = count;
+ opp->clks = devm_kzalloc(dev, sizeof(struct clk *) * count, GFP_KERNEL);
+ if (!opp->clks)
+ return ERR_PTR(-ENOMEM);
+
+ opp->rates = devm_kzalloc(dev, sizeof(unsigned long) * count, GFP_KERNEL);
+ if (!opp->rates)
+ return ERR_PTR(-ENOMEM);
+
+ count = 0;
+ list_for_each_entry(icc_node, &provider->nodes, node_list)
+ count++;
+
+ opp->nodes = devm_kzalloc(dev, sizeof(*opp->nodes) * count, GFP_KERNEL);
+ if (!opp->nodes)
+ return ERR_PTR(-ENOMEM);
+ opp->nodes_count = count;
+
+ return opp;
+}
+
+static int busfreq_init_opp(struct icc_provider *provider,
+ struct busfreq_opp *opp,
+ struct busfreq_plat_opp *plat_opp)
+{
+ struct busfreq_icc_desc *desc = provider->data;
+ struct device *dev = desc->dev;
+ struct busfreq_opp_node *node;
+ struct icc_node *icc_node;
+ int i, j;
+
+ opp->perf_level = 0;
+ for (i = 0; i < opp->clks_count; i++) {
+ opp->clks[i] = devm_clk_get(dev, plat_opp->clks[i].name);
+ if (IS_ERR(opp->clks[i])) {
+ dev_err(dev, "Failed to get clock %s\n",
+ plat_opp->clks[i].name);
+ return PTR_ERR(opp->clks[i]);
+ }
+ opp->rates[i] = plat_opp->clks[i].rate;
+ opp->perf_level += opp->rates[i] >> 10;
+ }
+
+ i = 0;
+ list_for_each_entry(icc_node, &provider->nodes, node_list) {
+ node = &opp->nodes[i++];
+ node->icc_node = icc_node;
+ }
+
+ for (i = 0; i < plat_opp->nodes_count; i++) {
+ icc_node = busfreq_icc_node_get(provider,
+ plat_opp->nodes[i].id);
+ if (IS_ERR_OR_NULL(icc_node))
+ return -EINVAL;
+
+ for (j = 0, node = &opp->nodes[j]; j < opp->nodes_count;
+ j++, node = &opp->nodes[j]) {
+ if (node->icc_node == icc_node) {
+ node->min_avg_bw = BUSFREQ_UNDEFINED_BW;
+ node->min_peak_bw = BUSFREQ_UNDEFINED_BW;
+ }
+ }
+ }
+
+ INIT_LIST_HEAD(&opp->node);
+
+ return 0;
+}
+
+static int busfreq_register_opps(struct device *dev,
+ struct icc_provider *provider,
+ struct busfreq_plat_opp *busfreq_opps,
+ int count)
+{
+ struct busfreq_icc_desc *desc = provider->data;
+ struct busfreq_opp *opp;
+ int ret;
+ int i;
+
+ for (i = 0; i < count; i++) {
+ opp = busfreq_opp_alloc(provider, busfreq_opps[i].clks_count);
+ if (IS_ERR(opp))
+ return PTR_ERR(opp);
+
+ ret = busfreq_init_opp(provider, opp, &busfreq_opps[i]);
+ if (ret)
+ return ret;
+
+ if (busfreq_opps[i].default_opp)
+ desc->default_opp = opp;
+
+ list_add(&opp->node, &desc->opps);
+ }
+
+ return 0;
+}
+
+static void busfreq_unregister_opps(struct icc_provider *provider)
+{
+ struct busfreq_icc_desc *desc = provider->data;
+ struct device *dev = desc->dev;
+ struct busfreq_opp *opp, *tmp_opp;
+
+ list_for_each_entry_safe(opp, tmp_opp, &desc->opps, node) {
+ list_del(&opp->node);
+ devm_kfree(dev, opp->nodes);
+ devm_kfree(dev, opp->clks);
+ devm_kfree(dev, opp->rates);
+ devm_kfree(dev, opp);
+ }
+}
+
+int busfreq_register(struct platform_device *pdev,
+ struct busfreq_icc_node *busfreq_nodes, int nodes_count,
+ struct busfreq_plat_opp *busfreq_opps, int opps_count)
+{
+ struct device *dev = &pdev->dev;
+ struct busfreq_icc_desc *desc;
+ struct icc_provider *provider;
+ int ret;
+
+ desc = devm_kzalloc(dev, sizeof(*desc), GFP_KERNEL);
+ if (!desc)
+ return -ENOMEM;
+ desc->dev = dev;
+ mutex_init(&desc->mutex);
+ INIT_LIST_HEAD(&desc->opps);
+
+ provider = devm_kzalloc(dev, sizeof(*provider), GFP_KERNEL);
+ if (!provider)
+ return -ENOMEM;
+ provider->set = busfreq_icc_set;
+ provider->aggregate = busfreq_icc_aggregate;
+ provider->xlate = busfreq_icc_xlate;
+ provider->data = desc;
+ provider->dev = dev;
+ platform_set_drvdata(pdev, provider);
+
+ ret = icc_provider_add(provider);
+ if (ret) {
+ dev_err(dev, "error adding interconnect provider\n");
+ return ret;
+ }
+
+ ret = busfreq_register_nodes(provider, busfreq_nodes, nodes_count);
+ if (ret) {
+ dev_err(dev, "error adding interconnect nodes\n");
+ goto provider_del;
+ }
+
+ ret = busfreq_register_opps(dev, provider, busfreq_opps, opps_count);
+ if (ret) {
+ dev_err(dev, "error adding busfreq opp\n");
+ goto unregister_nodes;
+ }
+
+ ret = busfreq_set_opp(desc, desc->default_opp);
+ if (ret)
+ goto unregister_opps;
+
+ desc->pm_notifier.notifier_call = busfreq_pm_notify;
+ register_pm_notifier(&desc->pm_notifier);
+
+ desc->reboot_notifier.notifier_call = busfreq_reboot_notify;
+ register_reboot_notifier(&desc->reboot_notifier);
+
+ return 0;
+
+unregister_opps:
+ busfreq_unregister_opps(provider);
+unregister_nodes:
+ busfreq_unregister_nodes(provider);
+provider_del:
+ icc_provider_del(provider);
+ return ret;
+}
+EXPORT_SYMBOL_GPL(busfreq_register);
+
+int busfreq_unregister(struct platform_device *pdev)
+{
+ struct icc_provider *provider = platform_get_drvdata(pdev);
+ struct busfreq_icc_desc *desc = provider->data;
+ int ret;
+
+ unregister_reboot_notifier(&desc->reboot_notifier);
+ unregister_pm_notifier(&desc->pm_notifier);
+
+ ret = busfreq_set_opp(desc, desc->default_opp);
+ if (ret)
+ return ret;
+
+ icc_provider_del(provider);
+
+ busfreq_unregister_opps(provider);
+ busfreq_unregister_nodes(provider);
+ devm_kfree(&pdev->dev, desc);
+ devm_kfree(&pdev->dev, provider);
+
+ return 0;
+}
+EXPORT_SYMBOL_GPL(busfreq_unregister);
diff --git a/drivers/interconnect/imx/busfreq.h b/drivers/interconnect/imx/busfreq.h
new file mode 100644
index 000000000000..a60481f10500
--- /dev/null
+++ b/drivers/interconnect/imx/busfreq.h
@@ -0,0 +1,123 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Interconnect framework driver for i.MX SoC
+ *
+ * Copyright (c) 2019, BayLibre
+ * Author: Alexandre Bailon <abailon@baylibre.com>
+ */
+#ifndef __BUSFREQ_H
+#define __BUSFREQ_H
+
+#include <linux/kernel.h>
+
+#define BUSFREQ_MAX_LINKS 32
+#define BUSFREQ_UNDEFINED_BW 0xffffffff
+
+/*
+ * struct busfreq_icc_node - Describe an interconnect node
+ * @name: name of the node
+ * @id: an unique id to identify the node
+ * @links: an array of slaves' node id
+ * @num_links: number of id defined in links
+ */
+struct busfreq_icc_node {
+ char *name;
+ u16 id;
+ u16 links[BUSFREQ_MAX_LINKS];
+ u16 num_links;
+};
+
+/*
+ * struct busfreq_opp_clk - Clock name and rate to set for an opp
+ * @name: name of clock
+ * @rate: the rate to set when the opp is enabled
+ */
+struct busfreq_opp_clk {
+ char *name;
+ unsigned long rate;
+};
+
+/*
+ * struct busfreq_opp_bw - Describe a condition to meet to enable an opp
+ * @id: id of the node to test
+ * @avg_bw: minimum average bandwidth required to enable the opp, or
+ * ignored if set to BUSFREQ_UNDEFINED_BW
+ * @peak_bw: minimum peak bandwidth required to enable the opp, or
+ * ignored if set to BUSFREQ_UNDEFINED_BW
+ */
+struct busfreq_opp_bw {
+ u16 id;
+ u32 avg_bw;
+ u32 peak_bw;
+};
+
+/*
+ * struct busfreq_plat_opp - Describe an opp to register in busfreq
+ * @clks: array of clocks to configure when the opp is enable
+ * @clks_count: number of clocks
+ * @nodes: array of opp nodes (condition to meet for each node to enable opp)
+ * @nodes_count: number of nodes
+ * @default_opp: use this opp as default opp if true
+ */
+struct busfreq_plat_opp {
+ struct busfreq_opp_clk *clks;
+ int clks_count;
+ struct busfreq_opp_bw *nodes;
+ int nodes_count;
+ bool default_opp;
+};
+
+#define DEFINE_BUS_INTERCONNECT(_name, _id, _numlinks, ...) \
+ { \
+ .id = _id, \
+ .name = _name, \
+ .num_links = _numlinks, \
+ .links = { __VA_ARGS__ }, \
+ }
+
+#define DEFINE_BUS_MASTER(_name, _id, _dest_id) \
+ DEFINE_BUS_INTERCONNECT(_name, _id, 1, _dest_id)
+
+#define DEFINE_BUS_SLAVE(_name, _id) \
+ DEFINE_BUS_INTERCONNECT(_name, _id, 0)
+
+#define DEFINE_OPP_CLOCK(_name, _rate) \
+ { \
+ .name = _name, \
+ .rate = _rate, \
+ }
+
+#define DEFINE_OPP_BW(_id, _avg, _peak) \
+ { \
+ .id = _id, \
+ .avg_bw = _avg, \
+ .peak_bw = _peak, \
+ }
+
+#define DEFINE_OPP_NODE(_id) \
+ DEFINE_OPP_BW(_id, BUSFREQ_UNDEFINED_BW, BUSFREQ_UNDEFINED_BW)
+
+#define DEFINE_OPP(_clks, _nodes, _default) \
+ { \
+ .clks = _clks, \
+ .clks_count = ARRAY_SIZE(_clks), \
+ .nodes = _nodes, \
+ .nodes_count = ARRAY_SIZE(_nodes), \
+ .default_opp = _default, \
+ }
+
+#define DEFINE_OPP_NO_NODES(_clks, _default) \
+ { \
+ .clks = _clks, \
+ .clks_count = ARRAY_SIZE(_clks), \
+ .nodes = NULL, \
+ .nodes_count = 0, \
+ .default_opp = _default, \
+ }
+
+int busfreq_register(struct platform_device *pdev,
+ struct busfreq_icc_node *busfreq_nodes, int nodes_count,
+ struct busfreq_plat_opp *busfreq_opps, int opps_count);
+int busfreq_unregister(struct platform_device *pdev);
+
+#endif /* __BUSFREQ_H */
--
2.17.1
^ permalink raw reply related
* [RFCv2 6/8] devfreq: Add imx-devfreq driver
From: Leonard Crestez @ 2019-06-28 7:39 UTC (permalink / raw)
To: Alexandre Bailon, Georgi Djakov, Stephen Boyd, Michael Turquette,
Viresh Kumar
Cc: MyungJoo Ham, Kyungmin Park, Shawn Guo, Dong Aisheng,
Fabio Estevam, Rafael J. Wysocki, Jacky Bai, Anson Huang,
Abel Vesa, Krzysztof Kozlowski, Ulf Hansson, Saravana Kannan,
kernel, linux-imx, linux-pm, linux-clk, linux-arm-kernel
In-Reply-To: <cover.1561707104.git.leonard.crestez@nxp.com>
Add initial support for frequency switches on pieces of the imx
interconnect fabric.
Uses clk_set_min_rate so that other subsytems can also impose minimum
rate requests.
Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com>
---
drivers/devfreq/Kconfig | 10 +++
drivers/devfreq/Makefile | 1 +
drivers/devfreq/imx-devfreq.c | 142 ++++++++++++++++++++++++++++++++++
3 files changed, 153 insertions(+)
create mode 100644 drivers/devfreq/imx-devfreq.c
diff --git a/drivers/devfreq/Kconfig b/drivers/devfreq/Kconfig
index f3b242987fd9..3a7c6265ca2f 100644
--- a/drivers/devfreq/Kconfig
+++ b/drivers/devfreq/Kconfig
@@ -90,10 +90,20 @@ config ARM_EXYNOS_BUS_DEVFREQ
Each memory bus group could contain many memoby bus block. It reads
PPMU counters of memory controllers by using DEVFREQ-event device
and adjusts the operating frequencies and voltages with OPP support.
This does not yet operate with optimal voltages.
+config ARM_IMX_DEVFREQ
+ tristate "i.MX DEVFREQ Driver"
+ depends on ARCH_MXC
+ select DEVFREQ_GOV_USERSPACE
+ select PM_OPP
+ help
+ This adds the DEVFREQ driver for the Tegra family of SoCs.
+ It reads ACTMON counters of memory controllers and adjusts the
+ operating frequencies and voltages with OPP support.
+
config ARM_TEGRA_DEVFREQ
tristate "NVIDIA Tegra30/114/124/210 DEVFREQ Driver"
depends on ARCH_TEGRA_3x_SOC || ARCH_TEGRA_114_SOC || \
ARCH_TEGRA_132_SOC || ARCH_TEGRA_124_SOC || \
ARCH_TEGRA_210_SOC || \
diff --git a/drivers/devfreq/Makefile b/drivers/devfreq/Makefile
index 338ae8440db6..c2463ed4c934 100644
--- a/drivers/devfreq/Makefile
+++ b/drivers/devfreq/Makefile
@@ -7,10 +7,11 @@ obj-$(CONFIG_DEVFREQ_GOV_POWERSAVE) += governor_powersave.o
obj-$(CONFIG_DEVFREQ_GOV_USERSPACE) += governor_userspace.o
obj-$(CONFIG_DEVFREQ_GOV_PASSIVE) += governor_passive.o
# DEVFREQ Drivers
obj-$(CONFIG_ARM_EXYNOS_BUS_DEVFREQ) += exynos-bus.o
+obj-$(CONFIG_ARM_IMX_DEVFREQ) += imx-devfreq.o
obj-$(CONFIG_ARM_RK3399_DMC_DEVFREQ) += rk3399_dmc.o
obj-$(CONFIG_ARM_TEGRA_DEVFREQ) += tegra30-devfreq.o
obj-$(CONFIG_ARM_TEGRA20_DEVFREQ) += tegra20-devfreq.o
# DEVFREQ Event Drivers
diff --git a/drivers/devfreq/imx-devfreq.c b/drivers/devfreq/imx-devfreq.c
new file mode 100644
index 000000000000..aa8868d32f22
--- /dev/null
+++ b/drivers/devfreq/imx-devfreq.c
@@ -0,0 +1,142 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2019 NXP
+ */
+
+#include <linux/clk.h>
+#include <linux/devfreq.h>
+#include <linux/devfreq-event.h>
+#include <linux/device.h>
+#include <linux/module.h>
+#include <linux/of_device.h>
+#include <linux/pm_opp.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+
+struct imx_devfreq {
+ struct devfreq_dev_profile profile;
+ struct devfreq *devfreq;
+ struct clk *clk;
+};
+
+static int imx_devfreq_target(struct device *dev, unsigned long *freq, u32 flags)
+{
+ struct imx_devfreq *priv = dev_get_drvdata(dev);
+ struct dev_pm_opp *new_opp;
+ unsigned long new_freq;
+ int ret;
+
+ new_opp = devfreq_recommended_opp(dev, freq, flags);
+ if (IS_ERR(new_opp)) {
+ ret = PTR_ERR(new_opp);
+ dev_err(dev, "failed to get recommended opp: %d\n", ret);
+ return ret;
+ }
+ new_freq = dev_pm_opp_get_freq(new_opp);
+ dev_pm_opp_put(new_opp);
+
+ ret = clk_set_min_rate(priv->clk, new_freq);
+ if (ret)
+ return ret;
+
+ ret = clk_set_rate(priv->clk, 0);
+ if (ret) {
+ clk_set_min_rate(priv->clk, priv->devfreq->previous_freq);
+ return ret;
+ }
+
+ return 0;
+}
+
+static int imx_devfreq_get_cur_freq(struct device *dev, unsigned long *freq)
+{
+ struct imx_devfreq *priv = dev_get_drvdata(dev);
+
+ *freq = clk_get_rate(priv->clk);
+
+ return 0;
+}
+
+static int imx_devfreq_get_dev_status(struct device *dev,
+ struct devfreq_dev_status *stat)
+{
+ struct imx_devfreq *priv = dev_get_drvdata(dev);
+
+ stat->busy_time = 0;
+ stat->total_time = 0;
+ stat->current_frequency = clk_get_rate(priv->clk);
+
+ return 0;
+}
+
+static void imx_devfreq_exit(struct device *dev)
+{
+ return dev_pm_opp_of_remove_table(dev);
+}
+
+static int imx_devfreq_probe(struct platform_device *pdev)
+{
+ struct device *dev = &pdev->dev;
+ struct imx_devfreq *priv;
+ int ret;
+
+ priv = devm_kzalloc(dev, sizeof(*priv), GFP_KERNEL);
+ if (!priv)
+ return -ENOMEM;
+
+ priv->clk = devm_clk_get(dev, NULL);
+ if (IS_ERR(priv->devfreq)) {
+ ret = PTR_ERR(priv->devfreq);
+ dev_err(dev, "failed to fetch clk: %d\n", ret);
+ return ret;
+ }
+ platform_set_drvdata(pdev, priv);
+
+ ret = dev_pm_opp_of_add_table(dev);
+ if (ret < 0) {
+ dev_err(dev, "failed to get OPP table\n");
+ return ret;
+ }
+
+ priv->profile.polling_ms = 1000;
+ priv->profile.target = imx_devfreq_target;
+ priv->profile.get_dev_status = imx_devfreq_get_dev_status;
+ priv->profile.exit = imx_devfreq_exit;
+ priv->profile.get_cur_freq = imx_devfreq_get_cur_freq;
+ priv->profile.initial_freq = clk_get_rate(priv->clk);
+
+ priv->devfreq = devm_devfreq_add_device(dev, &priv->profile,
+ DEVFREQ_GOV_USERSPACE,
+ NULL);
+ if (IS_ERR(priv->devfreq)) {
+ ret = PTR_ERR(priv->devfreq);
+ dev_err(dev, "failed to add devfreq device: %d\n", ret);
+ goto err_remove_table;
+ }
+
+ return 0;
+
+err_remove_table:
+ dev_pm_opp_of_remove_table(dev);
+ return ret;
+}
+
+static const struct of_device_id imx_devfreq_of_match[] = {
+ { .compatible = "fsl,imx8mm-ddrc", },
+ { .compatible = "fsl,imx8mm-noc", },
+ { .compatible = "fsl,imx8mm-nic", },
+ { /* sentinel */ },
+};
+MODULE_DEVICE_TABLE(of, imx_devfreq_of_match);
+
+static struct platform_driver imx_devfreq_platdrv = {
+ .probe = imx_devfreq_probe,
+ .driver = {
+ .name = "devfreq-imx",
+ .of_match_table = of_match_ptr(imx_devfreq_of_match),
+ },
+};
+module_platform_driver(imx_devfreq_platdrv);
+
+MODULE_DESCRIPTION("Generic i.MX bus frequency driver");
+MODULE_LICENSE("GPL v2");
--
2.17.1
^ permalink raw reply related
* [RFCv2 5/8] interconnect: imx: Add platform driver for imx8mm
From: Leonard Crestez @ 2019-06-28 7:39 UTC (permalink / raw)
To: Alexandre Bailon, Georgi Djakov, Stephen Boyd, Michael Turquette,
Viresh Kumar
Cc: MyungJoo Ham, Kyungmin Park, Shawn Guo, Dong Aisheng,
Fabio Estevam, Rafael J. Wysocki, Jacky Bai, Anson Huang,
Abel Vesa, Krzysztof Kozlowski, Ulf Hansson, Saravana Kannan,
kernel, linux-imx, linux-pm, linux-clk, linux-arm-kernel
In-Reply-To: <cover.1561707104.git.leonard.crestez@nxp.com>
From: Alexandre Bailon <abailon@baylibre.com>
This adds a platform driver for the i.MX8MM SoC.
Signed-off-by: Alexandre Bailon <abailon@baylibre.com>
Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com>
---
Changes by Leonard:
* Single DRAM clk
* Change compat string to imx8mm-interconnect
---
drivers/interconnect/imx/Kconfig | 4 +
drivers/interconnect/imx/Makefile | 1 +
drivers/interconnect/imx/busfreq-imx8mm.c | 151 ++++++++++++++++++++++
include/dt-bindings/interconnect/imx8mm.h | 49 +++++++
4 files changed, 205 insertions(+)
create mode 100644 drivers/interconnect/imx/busfreq-imx8mm.c
create mode 100644 include/dt-bindings/interconnect/imx8mm.h
diff --git a/drivers/interconnect/imx/Kconfig b/drivers/interconnect/imx/Kconfig
index afd7b71bbd82..b569d40e5ca0 100644
--- a/drivers/interconnect/imx/Kconfig
+++ b/drivers/interconnect/imx/Kconfig
@@ -9,5 +9,9 @@ config BUSFREQ
depends on INTERCONNECT_IMX
help
A generic interconnect driver that could be used for any i.MX.
This provides a way to register master and slave and some opp
to use when one or more master are in use.
+
+config BUSFREQ_IMX8MM
+ bool "i.MX8MM busfreq driver"
+ depends on BUSFREQ
diff --git a/drivers/interconnect/imx/Makefile b/drivers/interconnect/imx/Makefile
index fea647183815..a92fea6e042d 100644
--- a/drivers/interconnect/imx/Makefile
+++ b/drivers/interconnect/imx/Makefile
@@ -1 +1,2 @@
obj-$(CONFIG_BUSFREQ) += busfreq.o
+obj-$(CONFIG_BUSFREQ_IMX8MM) += busfreq-imx8mm.o
diff --git a/drivers/interconnect/imx/busfreq-imx8mm.c b/drivers/interconnect/imx/busfreq-imx8mm.c
new file mode 100644
index 000000000000..49ffe230637d
--- /dev/null
+++ b/drivers/interconnect/imx/busfreq-imx8mm.c
@@ -0,0 +1,151 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Interconnect framework driver for i.MX SoC
+ *
+ * Copyright (c) 2019, BayLibre
+ * Author: Alexandre Bailon <abailon@baylibre.com>
+ */
+
+#include <linux/device.h>
+#include <linux/module.h>
+#include <linux/of_platform.h>
+#include <linux/platform_device.h>
+
+#include <dt-bindings/interconnect/imx8mm.h>
+
+#include "busfreq.h"
+
+/*
+ * Describe bus masters, slaves and connections between them
+ *
+ * This is a simplified subset of the bus diagram, there are several other
+ * PL301 nics which are skipped/merged into PL301_MAIN
+ */
+static struct busfreq_icc_node imx8mm_icc_nodes[] = {
+ DEFINE_BUS_INTERCONNECT("NOC", IMX8MM_ICN_NOC, 2,
+ IMX8MM_ICS_DRAM,
+ IMX8MM_ICN_MAIN),
+
+ DEFINE_BUS_SLAVE("DRAM", IMX8MM_ICS_DRAM),
+ DEFINE_BUS_SLAVE("OCRAM", IMX8MM_ICS_OCRAM),
+ DEFINE_BUS_MASTER("A53", IMX8MM_ICM_A53, IMX8MM_ICN_NOC),
+
+ /* VPUMIX */
+ DEFINE_BUS_MASTER("VPU H1", IMX8MM_ICM_VPU_H1, IMX8MM_ICN_VIDEO),
+ DEFINE_BUS_MASTER("VPU G1", IMX8MM_ICM_VPU_G1, IMX8MM_ICN_VIDEO),
+ DEFINE_BUS_MASTER("VPU G2", IMX8MM_ICM_VPU_G2, IMX8MM_ICN_VIDEO),
+ DEFINE_BUS_INTERCONNECT("PL301_VIDEO", IMX8MM_ICN_VIDEO, 1, IMX8MM_ICN_NOC),
+
+ /* GPUMIX */
+ DEFINE_BUS_MASTER("GPU 2D", IMX8MM_ICM_GPU2D, IMX8MM_ICN_GPU),
+ DEFINE_BUS_MASTER("GPU 3D", IMX8MM_ICM_GPU3D, IMX8MM_ICN_GPU),
+ DEFINE_BUS_INTERCONNECT("PL301_GPU", IMX8MM_ICN_GPU, 1, IMX8MM_ICN_NOC),
+
+ /* DISPLAYMIX */
+ DEFINE_BUS_MASTER("CSI", IMX8MM_ICM_CSI, IMX8MM_ICN_MIPI),
+ DEFINE_BUS_MASTER("LCDIF", IMX8MM_ICM_LCDIF, IMX8MM_ICN_MIPI),
+ DEFINE_BUS_INTERCONNECT("PL301_MIPI", IMX8MM_ICN_MIPI, 1, IMX8MM_ICN_NOC),
+
+ /* HSIO */
+ DEFINE_BUS_MASTER("USB1", IMX8MM_ICM_USB1, IMX8MM_ICN_HSIO),
+ DEFINE_BUS_MASTER("USB2", IMX8MM_ICM_USB2, IMX8MM_ICN_HSIO),
+ DEFINE_BUS_MASTER("PCIE", IMX8MM_ICM_PCIE, IMX8MM_ICN_HSIO),
+ DEFINE_BUS_INTERCONNECT("PL301_HSIO", IMX8MM_ICN_HSIO, 1, IMX8MM_ICN_NOC),
+
+ /* Audio */
+ DEFINE_BUS_MASTER("SDMA2", IMX8MM_ICM_SDMA2, IMX8MM_ICN_AUDIO),
+ DEFINE_BUS_MASTER("SDMA3", IMX8MM_ICM_SDMA3, IMX8MM_ICN_AUDIO),
+ DEFINE_BUS_INTERCONNECT("PL301_AUDIO", IMX8MM_ICN_AUDIO, 1, IMX8MM_ICN_MAIN),
+
+ /* Ethernet */
+ DEFINE_BUS_MASTER("ENET", IMX8MM_ICM_ENET, IMX8MM_ICN_ENET),
+ DEFINE_BUS_INTERCONNECT("PL301_ENET", IMX8MM_ICN_ENET, 1, IMX8MM_ICN_MAIN),
+
+ /* Other */
+ DEFINE_BUS_MASTER("SDMA1", IMX8MM_ICM_SDMA1, IMX8MM_ICN_MAIN),
+ DEFINE_BUS_MASTER("NAND", IMX8MM_ICM_NAND, IMX8MM_ICN_MAIN),
+ DEFINE_BUS_MASTER("USDHC1", IMX8MM_ICM_USDHC1, IMX8MM_ICN_MAIN),
+ DEFINE_BUS_MASTER("USDHC2", IMX8MM_ICM_USDHC2, IMX8MM_ICN_MAIN),
+ DEFINE_BUS_MASTER("USDHC3", IMX8MM_ICM_USDHC3, IMX8MM_ICN_MAIN),
+ DEFINE_BUS_INTERCONNECT("PL301_MAIN", IMX8MM_ICN_MAIN, 2,
+ IMX8MM_ICN_NOC,
+ IMX8MM_ICS_OCRAM),
+};
+
+static struct busfreq_opp_clk imx8mm_low_freq_clks[] = {
+ DEFINE_OPP_CLOCK("dram", 25000000),
+ DEFINE_OPP_CLOCK("noc", 150000000),
+ DEFINE_OPP_CLOCK("ahb", 22222222),
+ DEFINE_OPP_CLOCK("axi", 24000000),
+};
+
+static struct busfreq_opp_clk imx8mm_audio_freq_clks[] = {
+ DEFINE_OPP_CLOCK("dram", 100000000),
+ DEFINE_OPP_CLOCK("noc", 150000000),
+ DEFINE_OPP_CLOCK("ahb", 22222222),
+ DEFINE_OPP_CLOCK("axi", 24000000),
+};
+
+static struct busfreq_opp_bw imx8mm_audio_freq_nodes[] = {
+ DEFINE_OPP_NODE(IMX8MM_ICM_SDMA2),
+ DEFINE_OPP_NODE(IMX8MM_ICM_SDMA3),
+};
+
+static struct busfreq_opp_clk imx8mm_high_freq_clks[] = {
+ DEFINE_OPP_CLOCK("dram", 750000000),
+ DEFINE_OPP_CLOCK("noc", 750000000),
+ DEFINE_OPP_CLOCK("ahb", 133333333),
+ DEFINE_OPP_CLOCK("axi", 333333333),
+};
+
+static struct busfreq_opp_bw imx8mm_high_freq_nodes[] = {
+ DEFINE_OPP_NODE(IMX8MM_ICM_SDMA2),
+ DEFINE_OPP_NODE(IMX8MM_ICM_SDMA3),
+ DEFINE_OPP_NODE(IMX8MM_ICM_VPU_H1),
+ DEFINE_OPP_NODE(IMX8MM_ICM_VPU_G1),
+ DEFINE_OPP_NODE(IMX8MM_ICM_VPU_G2),
+ DEFINE_OPP_NODE(IMX8MM_ICM_GPU2D),
+ DEFINE_OPP_NODE(IMX8MM_ICM_GPU3D),
+ DEFINE_OPP_NODE(IMX8MM_ICM_CSI),
+ DEFINE_OPP_NODE(IMX8MM_ICM_LCDIF),
+};
+
+static struct busfreq_plat_opp imx8mm_opps[] = {
+ DEFINE_OPP_NO_NODES(imx8mm_low_freq_clks, false),
+ DEFINE_OPP(imx8mm_audio_freq_clks, imx8mm_audio_freq_nodes, false),
+ DEFINE_OPP(imx8mm_high_freq_clks, imx8mm_high_freq_nodes, true),
+};
+
+static int imx8mm_busfreq_probe(struct platform_device *pdev)
+{
+ int ret;
+
+ ret = busfreq_register(pdev, imx8mm_icc_nodes,
+ ARRAY_SIZE(imx8mm_icc_nodes),
+ imx8mm_opps, ARRAY_SIZE(imx8mm_opps));
+ return ret;
+}
+
+static int imx8mm_busfreq_remove(struct platform_device *pdev)
+{
+ return busfreq_unregister(pdev);
+}
+
+static const struct of_device_id busfreq_of_match[] = {
+ { .compatible = "fsl,imx8mm-interconnect" },
+ { },
+};
+MODULE_DEVICE_TABLE(of, busfreq_of_match);
+
+static struct platform_driver imx8mm_busfreq_driver = {
+ .probe = imx8mm_busfreq_probe,
+ .remove = imx8mm_busfreq_remove,
+ .driver = {
+ .name = "imx8mm-interconnect",
+ .of_match_table = busfreq_of_match,
+ },
+};
+
+builtin_platform_driver(imx8mm_busfreq_driver);
+MODULE_AUTHOR("Alexandre Bailon <abailon@baylibre.com>");
+MODULE_LICENSE("GPL v2");
diff --git a/include/dt-bindings/interconnect/imx8mm.h b/include/dt-bindings/interconnect/imx8mm.h
new file mode 100644
index 000000000000..293a99cbc1af
--- /dev/null
+++ b/include/dt-bindings/interconnect/imx8mm.h
@@ -0,0 +1,49 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Interconnect framework driver for i.MX SoC
+ *
+ * Copyright (c) 2019, BayLibre
+ * Author: Alexandre Bailon <abailon@baylibre.com>
+ */
+
+#ifndef __IMX8MM_ICM_INTERCONNECT_IDS_H
+#define __IMX8MM_ICM_INTERCONNECT_IDS_H
+
+#define IMX8MM_ICN_NOC 1
+#define IMX8MM_ICS_DRAM 2
+#define IMX8MM_ICS_OCRAM 3
+#define IMX8MM_ICM_A53 4
+
+#define IMX8MM_ICM_VPU_H1 5
+#define IMX8MM_ICM_VPU_G1 6
+#define IMX8MM_ICM_VPU_G2 7
+#define IMX8MM_ICN_VIDEO 8
+
+#define IMX8MM_ICM_GPU2D 9
+#define IMX8MM_ICM_GPU3D 10
+#define IMX8MM_ICN_GPU 11
+
+#define IMX8MM_ICM_CSI 12
+#define IMX8MM_ICM_LCDIF 13
+#define IMX8MM_ICN_MIPI 14
+
+#define IMX8MM_ICM_USB1 11
+#define IMX8MM_ICM_USB2 12
+#define IMX8MM_ICM_PCIE 13
+#define IMX8MM_ICN_HSIO 14
+
+#define IMX8MM_ICM_SDMA2 15
+#define IMX8MM_ICM_SDMA3 16
+#define IMX8MM_ICN_AUDIO 17
+
+#define IMX8MM_ICM_ENET 18
+#define IMX8MM_ICN_ENET 19
+
+#define IMX8MM_ICN_MAIN 20
+#define IMX8MM_ICM_NAND 21
+#define IMX8MM_ICM_SDMA1 22
+#define IMX8MM_ICM_USDHC1 23
+#define IMX8MM_ICM_USDHC2 24
+#define IMX8MM_ICM_USDHC3 25
+
+#endif /* __IMX8MM_ICM_INTERCONNECT_IDS_H */
--
2.17.1
^ permalink raw reply related
* [RFCv2 3/8] arm64: dts: imx8mm: Add dram dvfs irqs to ccm node
From: Leonard Crestez @ 2019-06-28 7:39 UTC (permalink / raw)
To: Alexandre Bailon, Georgi Djakov, Stephen Boyd, Michael Turquette,
Viresh Kumar
Cc: MyungJoo Ham, Kyungmin Park, Shawn Guo, Dong Aisheng,
Fabio Estevam, Rafael J. Wysocki, Jacky Bai, Anson Huang,
Abel Vesa, Krzysztof Kozlowski, Ulf Hansson, Saravana Kannan,
kernel, linux-imx, linux-pm, linux-clk, linux-arm-kernel
In-Reply-To: <cover.1561707104.git.leonard.crestez@nxp.com>
This could probably be avoided by handling these in secure world.
Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com>
---
arch/arm64/boot/dts/freescale/imx8mm.dtsi | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/arch/arm64/boot/dts/freescale/imx8mm.dtsi b/arch/arm64/boot/dts/freescale/imx8mm.dtsi
index 232a7412755a..5da905c257ad 100644
--- a/arch/arm64/boot/dts/freescale/imx8mm.dtsi
+++ b/arch/arm64/boot/dts/freescale/imx8mm.dtsi
@@ -449,10 +449,18 @@
#clock-cells = <1>;
clocks = <&osc_32k>, <&osc_24m>, <&clk_ext1>, <&clk_ext2>,
<&clk_ext3>, <&clk_ext4>;
clock-names = "osc_32k", "osc_24m", "clk_ext1", "clk_ext2",
"clk_ext3", "clk_ext4";
+ interrupts = <GIC_SPI 74 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 75 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 76 IRQ_TYPE_LEVEL_HIGH>,
+ <GIC_SPI 77 IRQ_TYPE_LEVEL_HIGH>;
+ interrupt-name = "irq_busfreq_0",
+ "irq_busfreq_1",
+ "irq_busfreq_2",
+ "irq_busfreq_3";
};
src: reset-controller@30390000 {
compatible = "fsl,imx8mm-src", "syscon";
reg = <0x30390000 0x10000>;
--
2.17.1
^ permalink raw reply related
* [RFCv2 2/8] clk: imx8m-composite: Switch to determine_rate
From: Leonard Crestez @ 2019-06-28 7:39 UTC (permalink / raw)
To: Alexandre Bailon, Georgi Djakov, Stephen Boyd, Michael Turquette,
Viresh Kumar
Cc: MyungJoo Ham, Kyungmin Park, Shawn Guo, Dong Aisheng,
Fabio Estevam, Rafael J. Wysocki, Jacky Bai, Anson Huang,
Abel Vesa, Krzysztof Kozlowski, Ulf Hansson, Saravana Kannan,
kernel, linux-imx, linux-pm, linux-clk, linux-arm-kernel
In-Reply-To: <cover.1561707104.git.leonard.crestez@nxp.com>
This allows consumers to use min_rate max_rate.
Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com>
---
drivers/clk/imx/clk-composite-8m.c | 34 +++++++++++++++++++-----------
1 file changed, 22 insertions(+), 12 deletions(-)
diff --git a/drivers/clk/imx/clk-composite-8m.c b/drivers/clk/imx/clk-composite-8m.c
index 388bdb94f841..1be82ec08ecd 100644
--- a/drivers/clk/imx/clk-composite-8m.c
+++ b/drivers/clk/imx/clk-composite-8m.c
@@ -45,10 +45,12 @@ static unsigned long imx8m_clk_composite_divider_recalc_rate(struct clk_hw *hw,
divider->flags, PCG_DIV_WIDTH);
}
static int imx8m_clk_composite_compute_dividers(unsigned long rate,
unsigned long parent_rate,
+ unsigned long min_rate,
+ unsigned long max_rate,
int *prediv, int *postdiv)
{
int div1, div2;
int error = INT_MAX;
int ret = -EINVAL;
@@ -56,11 +58,17 @@ static int imx8m_clk_composite_compute_dividers(unsigned long rate,
*prediv = 1;
*postdiv = 1;
for (div1 = 1; div1 <= PCG_PREDIV_MAX; div1++) {
for (div2 = 1; div2 <= PCG_DIV_MAX; div2++) {
- int new_error = ((parent_rate / div1) / div2) - rate;
+ unsigned long new_rate;
+ int new_error;
+
+ new_rate = ((parent_rate / div1) / div2);
+ if (new_rate < min_rate || new_rate > max_rate)
+ continue;
+ new_error = new_rate - rate;
if (abs(new_error) < abs(error)) {
*prediv = div1;
*postdiv = div2;
error = new_error;
@@ -69,38 +77,40 @@ static int imx8m_clk_composite_compute_dividers(unsigned long rate,
}
}
return ret;
}
-static long imx8m_clk_composite_divider_round_rate(struct clk_hw *hw,
- unsigned long rate,
- unsigned long *prate)
+static int imx8m_clk_composite_divider_determine_rate(struct clk_hw *hw,
+ struct clk_rate_request *req)
{
int prediv_value;
int div_value;
- imx8m_clk_composite_compute_dividers(rate, *prate,
- &prediv_value, &div_value);
- rate = DIV_ROUND_UP(*prate, prediv_value);
+ imx8m_clk_composite_compute_dividers(req->rate, req->best_parent_rate,
+ req->min_rate, req->max_rate,
+ &prediv_value, &div_value);
- return DIV_ROUND_UP(rate, div_value);
+ req->rate = DIV_ROUND_UP(req->best_parent_rate, prediv_value);
+ req->rate = DIV_ROUND_UP(req->rate, div_value);
+ return 0;
}
static int imx8m_clk_composite_divider_set_rate(struct clk_hw *hw,
- unsigned long rate,
- unsigned long parent_rate)
+ unsigned long rate,
+ unsigned long parent_rate)
{
struct clk_divider *divider = to_clk_divider(hw);
unsigned long flags = 0;
int prediv_value;
int div_value;
int ret;
u32 val;
ret = imx8m_clk_composite_compute_dividers(rate, parent_rate,
- &prediv_value, &div_value);
+ 0, ULONG_MAX,
+ &prediv_value, &div_value);
if (ret)
return -EINVAL;
spin_lock_irqsave(divider->lock, flags);
@@ -117,11 +127,11 @@ static int imx8m_clk_composite_divider_set_rate(struct clk_hw *hw,
return ret;
}
static const struct clk_ops imx8m_clk_composite_divider_ops = {
.recalc_rate = imx8m_clk_composite_divider_recalc_rate,
- .round_rate = imx8m_clk_composite_divider_round_rate,
+ .determine_rate = imx8m_clk_composite_divider_determine_rate,
.set_rate = imx8m_clk_composite_divider_set_rate,
};
struct clk *imx8m_clk_composite_flags(const char *name,
const char * const *parent_names,
--
2.17.1
^ permalink raw reply related
* [RFCv2 1/8] clk: imx8mm: Add dram freq switch support
From: Leonard Crestez @ 2019-06-28 7:39 UTC (permalink / raw)
To: Alexandre Bailon, Georgi Djakov, Stephen Boyd, Michael Turquette,
Viresh Kumar
Cc: MyungJoo Ham, Kyungmin Park, Shawn Guo, Dong Aisheng,
Fabio Estevam, Rafael J. Wysocki, Jacky Bai, Anson Huang,
Abel Vesa, Krzysztof Kozlowski, Ulf Hansson, Saravana Kannan,
kernel, linux-imx, linux-pm, linux-clk, linux-arm-kernel
In-Reply-To: <cover.1561707104.git.leonard.crestez@nxp.com>
Add a compound clock encapsulating dram frequency switch support for
imx8m chips. This allows higher-level DVFS code to manipulate dram
frequency using standard clock framework APIs.
Linux-side implementation is similar in principle to imx_clk_cpu or a
composite clock. Only some preparation is done inside the kernel, the
actual freq switch is performed from TF-A code which runs from an SRAM
area.
This is an early proof-of-concept which only support low/high mode on
imx8mm.
Signed-off-by: Leonard Crestez <leonard.crestez@nxp.com>
---
Changes since v1:
* Implement determin_rate so that consumers can set_min_rate
* Initial freq table
Link to v1:
Among the possible cleanups:
* Handle errors in low/high busfreq code and back off
* Move irq to secure world
* Try to use fewer clk parameters
* Use a table of frequencies
* More chips and frequencies
---
drivers/clk/imx/Makefile | 1 +
drivers/clk/imx/clk-imx8m-dram.c | 357 +++++++++++++++++++++++
drivers/clk/imx/clk-imx8mm.c | 12 +
drivers/clk/imx/clk.h | 13 +
include/dt-bindings/clock/imx8mm-clock.h | 4 +-
5 files changed, 386 insertions(+), 1 deletion(-)
create mode 100644 drivers/clk/imx/clk-imx8m-dram.c
diff --git a/drivers/clk/imx/Makefile b/drivers/clk/imx/Makefile
index 05641c64b317..0fc7195d6d3a 100644
--- a/drivers/clk/imx/Makefile
+++ b/drivers/clk/imx/Makefile
@@ -10,10 +10,11 @@ obj-$(CONFIG_MXC_CLK) += \
clk-fixup-div.o \
clk-fixup-mux.o \
clk-frac-pll.o \
clk-gate-exclusive.o \
clk-gate2.o \
+ clk-imx8m-dram.o \
clk-pfd.o \
clk-pfdv2.o \
clk-pllv1.o \
clk-pllv2.o \
clk-pllv3.o \
diff --git a/drivers/clk/imx/clk-imx8m-dram.c b/drivers/clk/imx/clk-imx8m-dram.c
new file mode 100644
index 000000000000..1d75be4a2f3a
--- /dev/null
+++ b/drivers/clk/imx/clk-imx8m-dram.c
@@ -0,0 +1,357 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * Copyright (c) 2019 NXP
+ */
+
+#define DEBUG
+#include <linux/arm-smccc.h>
+#include <linux/clk.h>
+#include <linux/clk-provider.h>
+#include <linux/device.h>
+#include <linux/interrupt.h>
+#include <linux/slab.h>
+#include <linux/of_irq.h>
+#include "clk.h"
+
+#define FSL_SIP_DDR_DVFS 0xc2000004
+
+/* Freq of setpoints is soc-dependent */
+#define FSL_SIP_DDR_FREQ_SET_HIGH 0x0
+#define FSL_SIP_DDR_FREQ_WAIT_DONE 0xf
+
+struct imx8m_dram_rate {
+ unsigned long rate;
+ unsigned int smcarg;
+};
+
+/*
+ * This clk roughly wraps the following clk structure:
+ *
+ * +----------+ |\ +------+
+ * | dram_pll |-------|M| dram_core | |
+ * +----------+ |U|---------->| D |
+ * /--|X| | D |
+ * dram_alt_root | |/ | R |
+ * | | C |
+ * +---------+ | |
+ * |FIX DIV/4| | |
+ * +---------+ | |
+ * composite: | | |
+ * +----------+ | | |
+ * | dram_alt |----/ | |
+ * +----------+ | |
+ * | dram_apb |-------------------->| |
+ * +----------+ +------+
+ *
+ * The DDR data rate is 4x dram_core
+ *
+ * The APB interface is only used for control registers and can otherwise
+ * be shut off.
+ *
+ * The dram_pll is used for higher rates and dram_alt is used for lower rates.
+ *
+ * The actual switch is done inside ATF, what this wrapper does is:
+ * - Enable the new parents
+ * - Call into ATF
+ * - Set the new rates
+ * - Set the new parents
+ * - Drop the reference count added to new parents at step 1
+ *
+ * In practice only 2 rates are supported: low and high.
+ */
+
+struct dram_clk {
+ struct clk_hw hw;
+ struct clk *dram_core;
+ struct clk *dram_apb;
+ struct clk *dram_pll;
+ struct clk *dram_alt;
+ struct clk *dram_alt_root;
+ struct clk *sys1_pll_40m;
+ struct clk *sys1_pll_100m;
+ struct clk *sys1_pll_800m;
+ int irqs[CONFIG_NR_CPUS];
+
+ unsigned int rate_count;
+ const struct imx8m_dram_rate *rate_table;
+};
+
+static inline struct dram_clk *to_dram_clk(struct clk_hw *hw)
+{
+ return container_of(hw, struct dram_clk, hw);
+}
+
+static irqreturn_t wait_in_wfe_irq(int irq, void *dev_id)
+{
+ struct arm_smccc_res res;
+
+ /* call smc trap to ATF */
+ arm_smccc_smc(FSL_SIP_DDR_DVFS, FSL_SIP_DDR_FREQ_WAIT_DONE, 0,
+ 0, 0, 0, 0, 0, &res);
+
+ return IRQ_HANDLED;
+}
+
+static void update_bus_freq(int target_freq)
+{
+ struct arm_smccc_res res;
+ u32 online_cpus = 0;
+ int cpu = 0;
+
+ local_irq_disable();
+
+ for_each_online_cpu(cpu)
+ online_cpus |= (1 << (cpu * 8));
+
+ /* change the ddr freqency */
+ arm_smccc_smc(FSL_SIP_DDR_DVFS, target_freq, online_cpus,
+ 0, 0, 0, 0, 0, &res);
+
+ local_irq_enable();
+}
+
+static int dram_clk_ensure_irq_affinity(struct dram_clk* priv)
+{
+ int err, cpu;
+
+ for_each_online_cpu(cpu) {
+ err = irq_set_affinity(priv->irqs[cpu], cpumask_of(cpu));
+ if (err) {
+ pr_err("imx8m_dram_clk set irqs[%d] affinity failed: %d\n",
+ cpu, err);
+ return err;
+ }
+ }
+
+ return 0;
+}
+
+/* Round UP */
+static const struct imx8m_dram_rate *dram_clk_find_rate(
+ struct dram_clk *priv,
+ unsigned long rate)
+{
+ int i;
+
+ for (i = priv->rate_count - 1; i >= 0; --i)
+ if (priv->rate_table[i].rate >= rate)
+ return &priv->rate_table[i];
+
+ return &priv->rate_table[0];
+}
+
+/* Round UP taking min and max into account */
+static int dram_clk_determine_rate(
+ struct clk_hw *hw,
+ struct clk_rate_request *req)
+{
+ struct dram_clk *priv = to_dram_clk(hw);
+ unsigned long tab_rate;
+ int i;
+
+ for (i = priv->rate_count - 1; i >= 0; --i) {
+ tab_rate = priv->rate_table[i].rate;
+ if (tab_rate >= req->rate &&
+ tab_rate >= req->min_rate &&
+ tab_rate <= req->max_rate)
+ {
+ req->rate = tab_rate;
+ return 0;
+ }
+ }
+
+ return -EINVAL;
+}
+
+static int dram_clk_set_low(struct dram_clk *priv)
+{
+ clk_prepare_enable(priv->sys1_pll_40m);
+ clk_prepare_enable(priv->dram_alt_root);
+ clk_prepare_enable(priv->sys1_pll_100m);
+
+ /* switch the DDR frequency */
+ update_bus_freq(0x2);
+
+ /* correct the clock tree info */
+ clk_set_parent(priv->dram_alt, priv->sys1_pll_100m);
+ clk_set_parent(priv->dram_core, priv->dram_alt_root);
+ clk_set_parent(priv->dram_apb, priv->sys1_pll_40m);
+ clk_set_rate(priv->dram_apb, 20000000);
+ clk_disable_unprepare(priv->sys1_pll_100m);
+ clk_disable_unprepare(priv->sys1_pll_40m);
+ clk_disable_unprepare(priv->dram_alt_root);
+ return 0;
+}
+
+static int dram_clk_set_high(struct dram_clk *priv)
+{
+ clk_prepare_enable(priv->sys1_pll_800m);
+ clk_prepare_enable(priv->dram_pll);
+
+ /* switch the DDR frequency */
+ update_bus_freq(0x0);
+
+ /* correct the clock tree info */
+ clk_set_parent(priv->dram_apb, priv->sys1_pll_800m);
+ clk_set_rate(priv->dram_apb, 160000000);
+ clk_set_parent(priv->dram_core, priv->dram_pll);
+ clk_disable_unprepare(priv->sys1_pll_800m);
+ clk_disable_unprepare(priv->dram_pll);
+
+ return 0;
+}
+
+static int dram_clk_set_rate(
+ struct clk_hw *hw,
+ unsigned long rate,
+ unsigned long parent_rate)
+{
+ struct dram_clk *priv = to_dram_clk(hw);
+ const struct imx8m_dram_rate *opp = dram_clk_find_rate(priv, rate);
+ int ret;
+
+ ret = dram_clk_ensure_irq_affinity(priv);
+ if (ret)
+ return ret;
+
+ if (opp->smcarg == 2) {
+ ret = dram_clk_set_low(priv);
+ } else if (opp->smcarg == 0) {
+ ret = dram_clk_set_high(priv);
+ } else {
+ ret = -EINVAL;
+ }
+
+ if (ret == 0)
+ pr_debug("%s freq set to %lu\n", clk_hw_get_name(hw), opp->rate);
+ else
+ pr_debug("%s freq set fail: %d\n", clk_hw_get_name(hw), ret);
+
+ return ret;
+}
+
+static unsigned long dram_clk_recalc_rate(struct clk_hw *hw, unsigned long parent_rate)
+{
+ struct dram_clk *priv = to_dram_clk(hw);
+
+ return clk_get_rate(priv->dram_core);
+}
+
+static const struct clk_ops dram_clk_ops = {
+ .determine_rate = dram_clk_determine_rate,
+ .recalc_rate = dram_clk_recalc_rate,
+ .set_rate = dram_clk_set_rate,
+};
+
+static int dram_clk_init_irqs(struct dram_clk* priv, struct device_node *np)
+{
+ int err, irq, cpu;
+
+ for_each_possible_cpu(cpu) {
+ irq = of_irq_get(np, cpu);
+ if (irq < 0) {
+ pr_err("imx8m_dram_clk fail of_irq_get %d\n", irq);
+ return irq;
+ }
+
+ err = request_irq(irq, wait_in_wfe_irq,
+ IRQF_PERCPU, "ddrc", NULL);
+ if (err) {
+ pr_err("imx8m_dram_clk request irq %d failed: %d\n",
+ irq, err);
+ return err;
+ }
+ priv->irqs[cpu] = irq;
+ }
+
+ return 0;
+}
+
+static void dram_clk_free_irqs(struct dram_clk* priv)
+{
+ int cpu;
+
+ for_each_possible_cpu(cpu) {
+ free_irq(priv->irqs[cpu], NULL);
+ priv->irqs[cpu] = 0;
+ }
+}
+
+static const struct imx8m_dram_rate imx8mq_dram_rate_table[] = {
+ {
+ .rate = 800000000,
+ .smcarg = 0x0,
+ },
+ {
+ .rate = 25000000,
+ .smcarg = 0x2,
+ },
+};
+
+static const struct imx8m_dram_rate imx8mm_dram_rate_table[] = {
+ {
+ .rate = 750000000,
+ .smcarg = 0x0,
+ },
+ {
+ .rate = 25000000,
+ .smcarg = 0x2,
+ },
+};
+
+struct clk* imx8m_dram_clk(
+ struct device_node* np,
+ const char *name, const char* parent_name,
+ struct clk* dram_core,
+ struct clk* dram_apb,
+ struct clk* dram_pll,
+ struct clk* dram_alt,
+ struct clk* dram_alt_root,
+ struct clk* sys1_pll_40m,
+ struct clk* sys1_pll_100m,
+ struct clk* sys1_pll_800m)
+{
+ struct dram_clk *priv;
+ struct clk *clk;
+ struct clk_init_data init;
+ int err;
+
+ priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+ if (!priv)
+ return ERR_PTR(-ENOMEM);
+
+ priv->dram_apb = dram_apb;
+ priv->dram_core = dram_core;
+ priv->dram_pll = dram_pll;
+ priv->dram_alt = dram_alt;
+ priv->dram_alt_root = dram_alt_root;
+ priv->sys1_pll_40m = sys1_pll_40m;
+ priv->sys1_pll_100m = sys1_pll_100m;
+ priv->sys1_pll_800m = sys1_pll_800m;
+ priv->rate_count = ARRAY_SIZE(imx8mm_dram_rate_table);
+ priv->rate_table = imx8mm_dram_rate_table;
+
+ init.name = name;
+ init.ops = &dram_clk_ops;
+ init.flags = CLK_IS_CRITICAL;
+ init.parent_names = &parent_name;
+ init.num_parents = 1;
+
+ err = dram_clk_init_irqs(priv, np);
+ if (err)
+ goto err_free_priv;
+
+ priv->hw.init = &init;
+ clk = clk_register(NULL, &priv->hw);
+ if (IS_ERR(clk)) {
+ err = PTR_ERR(clk);
+ goto err_free_irqs;
+ }
+ return clk;
+
+err_free_irqs:
+ dram_clk_free_irqs(priv);
+err_free_priv:
+ kfree(priv);
+ return ERR_PTR(err);
+}
diff --git a/drivers/clk/imx/clk-imx8mm.c b/drivers/clk/imx/clk-imx8mm.c
index 6b8e75df994d..8e19a6fdc415 100644
--- a/drivers/clk/imx/clk-imx8mm.c
+++ b/drivers/clk/imx/clk-imx8mm.c
@@ -660,10 +660,22 @@ static int __init imx8mm_clocks_init(struct device_node *ccm_node)
clks[IMX8MM_CLK_GPT_3M] = imx_clk_fixed_factor("gpt_3m", "osc_24m", 1, 8);
clks[IMX8MM_CLK_DRAM_ALT_ROOT] = imx_clk_fixed_factor("dram_alt_root", "dram_alt", 1, 4);
clks[IMX8MM_CLK_DRAM_CORE] = imx_clk_mux2_flags("dram_core_clk", base + 0x9800, 24, 1, imx8mm_dram_core_sels, ARRAY_SIZE(imx8mm_dram_core_sels), CLK_IS_CRITICAL);
+ clks[IMX8MM_CLK_DRAM] = imx8m_dram_clk(
+ ccm_node,
+ "dram", "dram_core_clk",
+ clks[IMX8MM_CLK_DRAM_CORE],
+ clks[IMX8MM_CLK_DRAM_APB],
+ clks[IMX8MM_DRAM_PLL_OUT],
+ clks[IMX8MM_CLK_DRAM_ALT],
+ clks[IMX8MM_CLK_DRAM_ALT_ROOT],
+ clks[IMX8MM_SYS_PLL1_40M],
+ clks[IMX8MM_SYS_PLL1_100M],
+ clks[IMX8MM_SYS_PLL1_800M]);
+
clks[IMX8MM_CLK_ARM] = imx_clk_cpu("arm", "arm_a53_div",
clks[IMX8MM_CLK_A53_DIV],
clks[IMX8MM_CLK_A53_SRC],
clks[IMX8MM_ARM_PLL_OUT],
clks[IMX8MM_CLK_24M]);
diff --git a/drivers/clk/imx/clk.h b/drivers/clk/imx/clk.h
index d94d9cb079d3..858547fbad58 100644
--- a/drivers/clk/imx/clk.h
+++ b/drivers/clk/imx/clk.h
@@ -468,6 +468,19 @@ struct clk *imx8m_clk_composite_flags(const char *name,
struct clk_hw *imx_clk_divider_gate(const char *name, const char *parent_name,
unsigned long flags, void __iomem *reg, u8 shift, u8 width,
u8 clk_divider_flags, const struct clk_div_table *table,
spinlock_t *lock);
+
+struct clk* imx8m_dram_clk(
+ struct device_node *np,
+ const char *name, const char* parent_name,
+ struct clk* dram_core,
+ struct clk* dram_apb,
+ struct clk* dram_pll,
+ struct clk* dram_alt,
+ struct clk* dram_alt_root,
+ struct clk* sys1_pll_40m,
+ struct clk* sys1_pll_100m,
+ struct clk* sys1_pll_800m);
+
#endif
diff --git a/include/dt-bindings/clock/imx8mm-clock.h b/include/dt-bindings/clock/imx8mm-clock.h
index 07e6c686f3ef..dde146b923a8 100644
--- a/include/dt-bindings/clock/imx8mm-clock.h
+++ b/include/dt-bindings/clock/imx8mm-clock.h
@@ -246,8 +246,10 @@
#define IMX8MM_CLK_GPIO5_ROOT 227
#define IMX8MM_CLK_SNVS_ROOT 228
#define IMX8MM_CLK_GIC 229
-#define IMX8MM_CLK_END 230
+#define IMX8MM_CLK_DRAM 230
+
+#define IMX8MM_CLK_END 231
#endif
--
2.17.1
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox