This is required to allow tasks to freely move between cgroups associated
with the tune controller.
Change-Id: I1f39b957462034586edc2fdc0a35488b314e9c8c
Signed-off-by: Syed Rameez Mustafa <rameezmustafa@codeaurora.org>
The schedtune controller will mimic the cpusets controller configuration
for now. For that we need to make 4 groups in addition to the root
group present by default.
Change-Id: I082f1e4e4ebf863e623cf66ee127eac70a3e2716
Signed-off-by: Syed Rameez Mustafa <rameezmustafa@codeaurora.org>
To support task performance boosting, the usage of a single knob has the
advantage to be a simple solution, both from the implementation and the
usability standpoint. However, on a real system it can be difficult to
identify a single value for the knob which fits the needs of multiple
different tasks. For example, some kernel threads and/or user-space
background services should be better managed the "standard" way while we
still want to be able to boost the performance of specific workloads.
In order to improve the flexibility of the task boosting mechanism this
patch is the first of a small series which extends the previous
implementation to introduce a "per task group" support.
This first patch introduces just the basic CGroups support, a new
"schedtune" CGroups controller is added which allows to configure
different boost value for different groups of tasks.
To keep the implementation simple but still effective for a boosting
strategy, the new controller:
1. allows only a two layer hierarchy
2. supports only a limited number of boost groups
A two layer hierarchy allows to place each task either:
a) in the root control group
thus being subject to a system-wide boosting value
b) in a child of the root group
thus being subject to the specific boost value defined by that
"boost group"
The limited number of "boost groups" supported is mainly motivated by
the observation that in a real system it could be useful to have only
few classes of tasks which deserve different treatment.
For example, background vs foreground or interactive vs low-priority.
As an additional benefit, a limited number of boost groups allows also
to have a simpler implementation especially for the code required to
compute the boost value for CPUs which have runnable tasks belonging to
different boost groups.
Change-Id: I1304e33a8440bfdad9c8bcf8129ff390216f2e32
cc: Tejun Heo <tj@kernel.org>
cc: Li Zefan <lizefan@huawei.com>
cc: Johannes Weiner <hannes@cmpxchg.org>
cc: Ingo Molnar <mingo@redhat.com>
cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Patrick Bellasi <patrick.bellasi@arm.com>
Git-commit: 13001f47c9
Git-repo: https://android.googlesource.com/kernel/common
Signed-off-by: Vikram Mulukutla <markivx@codeaurora.org>
Use the ddrss_gcc_dbg_clk mux index to get the rate
measurement of the DDR clock on MSMCOBALT. Note that
this feature is only supported on v2 platforms.
CRs-Fixed: 1068034
Change-Id: Ib472c226b7e35079e6b8391848a0a0d1c38962c0
Signed-off-by: Deepak Katragadda <dkatraga@codeaurora.org>
Battery temperature delta configuration can be used to know the
battery temperature change. There is an interrupt associated
with this configuration which can be used to update the battery
health of the device.
Change-Id: Ic55b9bb34ac9912f88dab6a11ec365814b525ca2
Signed-off-by: Subbaraman Narayanamurthy <subbaram@codeaurora.org>
GEN3 FG has jeita hysteresis support in hardware. Add support
to configure the hysteresis applied to jeita temperature via
a device tree property.
While at it, fix reading the JEITA thresholds from device tree
property where the total size of the elements was used instead
of the total number of elements.
Change-Id: I1d468f1291224de0f781ca71cbc1374a29d7c790
Signed-off-by: Subbaraman Narayanamurthy <subbaram@codeaurora.org>
When the battery is removed and a strong charger is plugged in,
device will be kept powered. However, because of missing battery
SRAM access over IMA can fail leading to errors. Eventually, the
errors will go away when the battery is inserted again. However,
if the user tries to access SRAM registers or bms power supply
properties, it can cause further errors flooding logs. Handle
this gracefully by returning an error upon battery missing is
detected.
CRs-Fixed: 1072401
Change-Id: I764442548cb076dad89856d75394ef9cf355f83a
Signed-off-by: Subbaraman Narayanamurthy <subbaram@codeaurora.org>
Open circuit voltage and predicted battery voltage are read from
16 bits. However, the real value is present only in 15 bits.
Mask the bit 16 to get the proper readings.
Change-Id: I196ccc64fd221fead87a88f43dce1a21ba800908
Signed-off-by: Subbaraman Narayanamurthy <subbaram@codeaurora.org>
Add a module parameter "fg_restart" to restart the fuel gauge
without loading the battery profile. This will be useful in
geting a fresh SOC estimate whenever required.
Change-Id: I0c054ad523a0eab72f777d010807bb3a55f63f7e
Signed-off-by: Subbaraman Narayanamurthy <subbaram@codeaurora.org>
Batteries age across the time. Capacity learning is a feature
which uses the hardware autonomous coulomb counter to determine
the actual battery capacity against the advertised full capacity.
This will be useful to determine how much degradation has
happened to the battery. Capacity learning algorithm will start
based on some pre-set conditions like start SOC, battery
temperature and ends when the charging cycle is complete.
Actual capacity will be exposed through charge_full property.
Change-Id: I89cc14d213b0de10bc8d052dd76e1468fe9c30e5
Signed-off-by: Subbaraman Narayanamurthy <subbaram@codeaurora.org>
Currently, FG_GEN3 driver loads the battery profile if it finds
out the profile loaded earlier doesn't match with the available
profile. This will be done when the profile integrity bit is
set already. Make this force profile loading as a configurable
parameter through a device tree property.
While at it, add a module parameter to dump the profile contents
conditionally. This will be helpful in debugging.
Change-Id: Ibaff55aa5434ee283e6d597485d193e9e52bb74e
Signed-off-by: Subbaraman Narayanamurthy <subbaram@codeaurora.org>
msoc-full is not used to determine when the battery is full. The charger
will wakeup the device when it has terminated charging indicating that
the battery is full. Disable the msoc-full IRQ as a wakeup source.
The msoc-delta IRQ is used to wakeup the device when the SOC changes
therefore having the bsoc-delta IRQ as a wakeup source is unnecessary.
Disable the bsoc-delta IRQ as a wakeup source.
Change-Id: Iab7087ce3d6727abc852d3a84590cfa570ec584d
Signed-off-by: Nicholas Troast <ntroast@codeaurora.org>
Currently, delta SOC threshold is written as a 8 bit value but
the encoding is in 11 bits. This will cause the delta SOC
threshold programmed to a lower value causing interrupts to fire
quickly. Change the numerator of delta SOC SRAM parameter so that
the delta SOC threshold will be set correctly. Maximum possible
threshold will be 12.45% (1/2048 * 255).
Change-Id: Ia908c917d71939b9d72b16ddca3400a692540bf3
Signed-off-by: Subbaraman Narayanamurthy <subbaram@codeaurora.org>
Configure CC_CV threshold voltage to 4.34V for both itech and
ascent battery profiles that will be used by GEN3 fuel gauge
on pmicobalt.
CRs-Fixed: 1068593
Change-Id: I7df3d343afa62ffdbf6f5db8c7fc7d94e2f0e3a0
Signed-off-by: Subbaraman Narayanamurthy <subbaram@codeaurora.org>
Based on the battery profile parameters, Vbatt greater than float
voltage and full voltage thresholds needs to be configured. Add
support for the same. Since this needs the battery profile
parameters to be read in advance before the hardware is
initialized, shuffle the code to get battery profile early.
CRs-Fixed: 1068593
Change-Id: If5df62d933cc653a276967b3d16c080c6d4ecb2e
Signed-off-by: Subbaraman Narayanamurthy <subbaram@codeaurora.org>
Fix the following issues found in static analysis:
- missing blank lines after declarations
- unnecessary return in a void function
- exporting world writable files is usually an error
- block comments use * on subsequent and trailing lines
- Add DT binding description and vendor prefixe for the
driver.
- Add Linux Foundation copyright markings.
Change-Id: Id2ed9a6d4e31bd2acd3185bf7ef9c78f8a41cf36
Signed-off-by: Himanshu Aggarwal <haggarwa@codeaurora.org>
Signed-off-by: Vevek Venkatesan <vevekv@codeaurora.org>
Signed-off-by: Shantanu Jain <shjain@codeaurora.org>
Change-Id: Ie366894e9e89707fa2950a7ffbfe4e93059a19a3
Signed-off-by: Lawrence Liao <lawrence_liao@pixart.com>
(cherry picked from commit dfc4349a8aa02bb0b7b0639ad5368a5c85f6520f)
Git-commit: dfc4349a8aa02bb0b7b0639ad5368a5c85f6520f
Git-repo: https://github.com/PixartOpen/pixart-sensor-db810-linux-driver
[shjain@codeaurora.org: removed below listed files from original
commit as they are not needed and to resolve trivial merge conflicts:
1. arch/arm/boot/dts/qcom/apq8094-dragonboard.dtsi
2. arch/arm/boot/dts/qcom/msm8994-pinctrl.dtsi
3. arch/arm/boot/dts/qcom/msm8994.dtsi
4. drivers/input/misc/Makefile
5. drivers/input/misc/ots_pat9125/Makefile
Remove the unused pieces of algorithm from the reference
driver of Pixart PAT9125 Rotating side button.]
Signed-off-by: Shantanu Jain <shjain@codeaurora.org>
Store the process id in the register trace buffer.
This gives a general idea about what process was last
running when the RTB stopped.
Change-Id: Idcbd8db9eb451f2d732c935f00b6c50f9af3c592
Signed-off-by: Se Wang (Patrick) Oh <sewango@codeaurora.org>
To support ADSP SVA on wcd934x, dedicated backend dailink needs
to be created. Add the dailink associating the specific cpu dai
and the MAD codec dai.
CRs-Fixed: 1070327
Change-Id: I4152d81e4e951216a8ecdd93562eb1740dbb962c
Signed-off-by: Walter Yang <yandongy@codeaurora.org>
Parse the Detailed Timing Descriptors of both EDID blocks for
resolutions. Check if the resolution is already added in local
data base. Also, add tolerance to fps and pclk variations for
a give resolution so that EDID parser can capture all the
resolutions correctly as published by the sink's EDID.
Change-Id: I113988eb21f82eef022723c421b1c1c9d4dead9f
Signed-off-by: Ajay Singh Parmar <aparmar@codeaurora.org>
