Pull module updates from Rusty Russell:
"Main excitement here is Peter Zijlstra's lockless rbtree optimization
to speed module address lookup. He found some abusers of the module
lock doing that too.
A little bit of parameter work here too; including Dan Streetman's
breaking up the big param mutex so writing a parameter can load
another module (yeah, really). Unfortunately that broke the usual
suspects, !CONFIG_MODULES and !CONFIG_SYSFS, so those fixes were
appended too"
* tag 'modules-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux: (26 commits)
modules: only use mod->param_lock if CONFIG_MODULES
param: fix module param locks when !CONFIG_SYSFS.
rcu: merge fix for Convert ACCESS_ONCE() to READ_ONCE() and WRITE_ONCE()
module: add per-module param_lock
module: make perm const
params: suppress unused variable error, warn once just in case code changes.
modules: clarify CONFIG_MODULE_COMPRESS help, suggest 'N'.
kernel/module.c: avoid ifdefs for sig_enforce declaration
kernel/workqueue.c: remove ifdefs over wq_power_efficient
kernel/params.c: export param_ops_bool_enable_only
kernel/params.c: generalize bool_enable_only
kernel/module.c: use generic module param operaters for sig_enforce
kernel/params: constify struct kernel_param_ops uses
sysfs: tightened sysfs permission checks
module: Rework module_addr_{min,max}
module: Use __module_address() for module_address_lookup()
module: Make the mod_tree stuff conditional on PERF_EVENTS || TRACING
module: Optimize __module_address() using a latched RB-tree
rbtree: Implement generic latch_tree
seqlock: Introduce raw_read_seqcount_latch()
...
Pull thermal management updates from Zhang Rui:
"Specifics:
- enhance Thermal Framework with several new capabilities:
* use power estimates
* compute weights with relative integers instead of percentages
* allow governors to have private data in thermal zones
* export thermal zone parameters through sysfs
Thanks to the ARM thermal team (Javi, Punit, KP).
- introduce a new thermal governor: power allocator. First in kernel
closed loop PI(D) controller for thermal control. Thanks to ARM
thermal team.
- enhance OF thermal to allow thermal zones to have sustainable power
HW specification. Thanks to Punit.
- introduce thermal driver for Intel Quark SoC x1000platform. Thanks
to Ong, Boon Leong.
- introduce QPNP PMIC temperature alarm driver. Thanks to Ivan T. I.
- introduce thermal driver for Hisilicon hi6220. Thanks to
kongxinwei.
- enhance Exynos thermal driver to handle Exynos5433 TMU. Thanks to
Chanwoo C.
- TI thermal driver now has a better implementation for EOCZ bit.
From Pavel M.
- add id for Skylake processors in int340x processor thermal driver.
- a couple of small fixes and cleanups."
* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/rzhang/linux: (36 commits)
thermal: hisilicon: add new hisilicon thermal sensor driver
dt-bindings: Document the hi6220 thermal sensor bindings
thermal: of-thermal: add support for reading coefficients property
thermal: support slope and offset coefficients
thermal: power_allocator: round the division when divvying up power
thermal: exynos: Add the support for Exynos5433 TMU
thermal: cpu_cooling: Fix power calculation when CPUs are offline
thermal: cpu_cooling: Remove cpu_dev update on policy CPU update
thermal: export thermal_zone_parameters to sysfs
thermal: cpu_cooling: Check memory allocation of power_table
ti-soc-thermal: request temperature periodically if hw can't do that itself
ti-soc-thermal: implement eocz bit to make driver useful on omap3
cleanup ti-soc-thermal
thermal: remove stale THERMAL_POWER_ACTOR select
thermal: Default OF created trip points to writable
thermal: core: Add Kconfig option to enable writable trips
thermal: x86_pkg_temp: drop const for thermal_zone_parameters
of: thermal: Introduce sustainable power for a thermal zone
thermal: add trace events to the power allocator governor
thermal: introduce the Power Allocator governor
...
This patch adds the support for hisilicon thermal sensor, within
hisilicon SoC. there will register sensors for thermal framework
and use device tree to bind cooling device.
Signed-off-by: Leo Yan <leo.yan@linaro.org>
Signed-off-by: kongxinwei <kong.kongxinwei@hisilicon.com>
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
Most code already uses consts for the struct kernel_param_ops,
sweep the kernel for the last offending stragglers. Other than
include/linux/moduleparam.h and kernel/params.c all other changes
were generated with the following Coccinelle SmPL patch. Merge
conflicts between trees can be handled with Coccinelle.
In the future git could get Coccinelle merge support to deal with
patch --> fail --> grammar --> Coccinelle --> new patch conflicts
automatically for us on patches where the grammar is available and
the patch is of high confidence. Consider this a feature request.
Test compiled on x86_64 against:
* allnoconfig
* allmodconfig
* allyesconfig
@ const_found @
identifier ops;
@@
const struct kernel_param_ops ops = {
};
@ const_not_found depends on !const_found @
identifier ops;
@@
-struct kernel_param_ops ops = {
+const struct kernel_param_ops ops = {
};
Generated-by: Coccinelle SmPL
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Junio C Hamano <gitster@pobox.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Kees Cook <keescook@chromium.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: cocci@systeme.lip6.fr
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Luis R. Rodriguez <mcgrof@suse.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
In order to avoid having each driver adding their own
specific DT property to specify slope and offset,
this patch adds a basic coefficient reading from
DT thermal zone node. Right now, as the thermal
framework does not support multiple sensors,
the current coefficients apply only to the only
sensor in the thermal zone.
The supported equation is a simple linear model:
slope * <sensor reading> + offset.
slope and offset are read from the coefficients
DT property. In the same way as it is described in
the DT thermal binding.
So, as of today, the thermal framework will support
only cases like:
/* hotspot = 1 * adc + 6000 */
coefficients = <1 6000>;
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
It is common to have a linear extrapolation from
the current sensor readings and the actual temperature
value. This is specially the case when the sensor
is in use to extrapolate hotspots.
This patch adds slope and offset constants for
single sensor linear extrapolation equation. Because
the same sensor can be use in different locations,
from board to board, these constants are added
as part of thermal_zone_params.
The constants are available through sysfs.
It is up to the device driver to determine
the usage of these values.
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
In situations where there is an uneven number of cooling devices, the
division of power among them can lead to a milliwatt being dropped on
the floor due to rounding errors. This doesn't sound like a lot, but
some devices only grant the lowest cooling device state for their
maximum power. So for instance, if the granted_power is the maximum
power and all devices are getting their maximum power, one would get
max_power - 1, making it choose cooling device state 1, instead of 0.
Round the division to make the calculation more accurate.
Cc: Zhang Rui <rui.zhang@intel.com>
Cc: Eduardo Valentin <edubezval@gmail.com>
Signed-off-by: Javi Merino <javi.merino@arm.com>
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
Package C8 to C10 was introduced in newer Intel CPUs, we need to
include them in the package c-state residency calculation.
Otherwise, idle injection target is not accurately maintained by
the closed control loop.
Also cleaned up the code to make it scale better with large number
of c-states.
Reported-by: Kristen Carlson Accardi <kristen@linux.intel.com>
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Broadwell server has support for package C-states, idle injection works
as expected on this platform.
Signed-off-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Mark the module init / exit functions with __init / __exit accodingly.
This allows making the intel_powerclamp_ids[] array __initconst, too, as
it only gets referenced from powerclamp_probe(). This is safe as
file2alias doesn't care about the section, but the symbol name for the
MODULE_DEVICE_TABLE alias.
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: Jacob Pan <jacob.jun.pan@linux.intel.com>
Signed-off-by: Mathias Krause <minipli@googlemail.com>
Acked-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Acked-by: Jacob Pan <jacob.jun.pan@linux.intel.com>
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
DESCRIPTION
Spurious Thermal Alert: Talert can happen randomly while the device remains
under the temperature limit defined for this event to trig. This spurious
event is caused by a incorrect re-synchronization between clock domains.
The comparison between configured threshold and current temperature value
can happen while the value is transitioning (metastable), thus causing
inappropriate event generation. No spurious event occurs as long as the
threshold value stays unchanged. Spurious event can be generated while a
thermal alert threshold is modified in
CONTROL_BANDGAP_THRESHOLD_MPU/GPU/CORE/DSPEVE/IVA_n.
WORKAROUND
Spurious event generation can be avoided by performing following sequence
when the threshold is modified:
1. Mask the hot/cold events at the thermal IP level.
2. Modify Threshold.
3. Unmask the hot/cold events at the thermal IP level.
Signed-off-by: Keerthy <j-keerthy@ti.com>
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
Bandgap Temperature read Dtemp can be corrupted
DESCRIPTION
Read accesses to registers listed below can be corrupted due to
incorrect resynchronization between clock domains.
Read access to registers below can be corrupted :
• CTRL_CORE_DTEMP_MPU/GPU/CORE/DSPEVE/IVA_n (n = 0 to 4)
• CTRL_CORE_TEMP_SENSOR_MPU/GPU/CORE/DSPEVE/IVA_n
WORKAROUND
Multiple reads to CTRL_CORE_TEMP_SENSOR_MPU/GPU/CORE/DSPEVE/IVA[9:0]:
BGAP_DTEMPMPU/GPU/CORE/DSPEVE/IVA is needed to discard false value and
read right value:
1. Perform two successive reads to BGAP_DTEMP bit field.
(a) If read1 returns Val1 and read2 returns Val1, then
right value is Val1.
(b) If read1 returns Val1, read 2 returns Val2, a third
read is needed.
2. Perform third read
(a) If read3 returns Val2 then right value is Val2.
(b) If read3 returns Val3, then right value is Val3.
The above in gist means if val1 and val2 are the same then we can go
ahead with that value else we need a third read which will be right
since synchronization will be complete by then.
Signed-off-by: Keerthy <j-keerthy@ti.com>
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
This patch adds the support for Exynos5433's TMU (Thermal Management Unit).
Exynos5433 has a little different register bit fields as following description:
- Support the eight trip points for rising/falling interrupt by using two registers
- Read the calibration type (1-point or 2-point) and sensor id from TRIMINFO register
- Use a little different register address
Cc: Zhang Rui <rui.zhang@intel.com>
Cc: Eduardo Valentin <edubezval@gmail.com>
Cc: Lukasz Majewski <l.majewski@samsung.com>
Signed-off-by: Chanwoo Choi <cw00.choi@samsung.com>
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
Ensure that the CPU for which the frequency is being requested
is online. If none of the CPUs are online the requested power is
returned as 0.
Acked-by: Javi Merino <javi.merino@arm.com>
Signed-off-by: Kapileshwar Singh <kapileshwar.singh@arm.com>
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
It was initially understood that an update to the cpu_device
(cached in cpufreq_cooling_device) was required to ascertain the
correct operating point of the device on a cpufreq policy->cpu update
or creation or deletion of a cpufreq policy.
(e.g. when the existing policy CPU goes offline).
This update is not required and it is possible to ascertain the OPPs
from the leading CPU in a cpufreq domain even if the CPU is hotplugged out.
Fixes: e0128d8ab423 ("thermal: cpu_cooling: implement the power cooling device API")
Acked-by: Javi Merino <javi.merino@arm.com>
Signed-off-by: Kapileshwar Singh <kapileshwar.singh@arm.com>
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
It's useful for tuning to be able to edit thermal_zone_parameters from
userspace. Export them to the thermal_zone sysfs so that they can be
easily changed.
Cc: Zhang Rui <rui.zhang@intel.com>
Cc: Eduardo Valentin <edubezval@gmail.com>
Signed-off-by: Javi Merino <javi.merino@arm.com>
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
We allocate the power_table in memory but we don't test whether the
allocation succeeded. Return -ENOMEM if kcalloc() fails.
Fixes: e0128d8ab423 ("thermal: cpu_cooling: implement the power cooling device API")
Cc: Eduardo Valentin <edubezval@gmail.com>
Cc: Zhang Rui <rui.zhang@intel.com>
Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Javi Merino <javi.merino@arm.com>
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
When periodic mode is not enabled, it is neccessary to force reads.
Signed-off-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
For omap3, proper implementation of eocz bit is needed. It was
actually a TODO in the driver.
Signed-off-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
Simplify code by removing goto's where they point to simple return.
Avoid confusing |= on error values.
Correct whitespace.
Signed-off-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Eduardo Valentin <edubezval@gmail.com>
