To have a full controll on smartphone, Installig custom rom like CyanogenMod is required. Installation Instructions :
- Download your proper version of CyanogenMod ( ROM )
- Intsalling Recovery (i.e. TWRP)
- Place ROM in your SD card or Internal Storage
- Crack and open bootloader
- Reboot into recovery mode
- Full wipe and factory reset (recommended)
- Back to main menu
- Flash ROM
- Reboot
After reboot go to Setting -> Developer Options -> toggle on root option.
First of all, this project was written in Java as Primary android programing language. Following Google's official support of Kotlin's programming language, the algorithm was overwritten with this new language. Since the Algorithm works with linux kernel and casue of C++ programming language speed, the algorithm was overwritten again. Speed of C++ helped us to achieve better performance and have a great results at benchmarks. In order to use C++ as android developing language, NDK which is a tool for compiling the codes to run on adnroid OS, is required. Another tools that should be downloaded in SDK are CMAKE and LLDB. Next step is configuring gradle. add the following code into your gradle.
cmake { cppFlags "" arguments "-DANDROID_STL=c++_static" }
You can use C++ in your application now.
There is a system file that you can configure Thermal Management setting, based on CyanogenMod kernel. As I used CyanogenMod 13.1 in Sony Xperia Z , the system file path was " /sys/module/msm_thermal/core_controll/enabled ". echo 0 to turn off and 1 to turn on the setting.
There are different system files to achieve cpu temperature:
- sys/class/thermal/thermal_zone0/temp
- sys/devices/virtual/thermal/thermal_zone0/temp
- sys/class/hwmon/hwmonX/temp1_input
You should notice that which thermal zone is relevant to temperature of cpu. so yoy should read all thermal_zones type. The paths is shown blow.
- sys/class/thermal/thermal_zone0/type
- sys/class/thermal/thermal_zone1/type
- sys/class/thermal/thermal_zone2/type
- .............
- .............
There are some ways to get cpu utilization. In this project the solution was running "top -m 1000 -n 1 -d 1" command in shell and calculate cpu usage via aggregate amounts in first line of result.