版权声明：本文为博主原创文章，未经博主允许不得转载。 https://blog.csdn.net/finewind/article/details/48732609
stop thermald
stop mpdecision
stop thermal-engine
# online A57
echo 1 > /sys/devices/system/cpu/cpu2/online
echo 1 > /sys/devices/system/cpu/cpu3/online
echo 1 > /sys/devices/system/cpu/cpu4/online
echo 1 > /sys/devices/system/cpu/cpu5/online
echo 1 > /sys/devices/system/cpu/cpu6/online
echo 1 > /sys/devices/system/cpu/cpu7/online
# set governor
echo performance > /sys/devices/system/cpu/cpu0/cpufreq/scaling_governor
echo performance > /sys/devices/system/cpu/cpu4/cpufreq/scaling_governor
# set max freq
echo 1555200 > /sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq
echo 1958400 > /sys/devices/system/cpu/cpu4/cpufreq/scaling_max_freq
# set gpu
echo performance > /sys/class/kgsl/kgsl-3d0/devfreq/governor
echo 0 > /sys/class/kgsl/kgsl-3d0/min_pwrlevel
echo performance > /sys/class/devfreq/qcom,cpubw.32/governor
echo performance > /sys/class/devfreq/qcom,gpubw.71/governor
echo performance > /sys/class/devfreq/qcom,kgsl-busmon.70/governorcpu0会一直在线，因此不用手动online。
8994的cpu频率是按簇调整的，因此只需要设置cpu0和cpu4的performance。
拉起来的瞬间可能会触发bcl的调频和拔核动作，全部拉起来后可能温度过高导致thermal降频及拔核。
另外，vendor/qcom/proprietary/android-perf/core-ctl/core_ctl.c的某些代码可能会引起拉核失败或者拉上来后又被这个模块给down了，需要做如下修改：
将753行cpu_callback()的CPU_UP_PREPARE分支里的如下判断改成if(0)，这可以防止拉核失败：
将该文件try_hotplug()函数里的do_hotplug()取消掉，可以防止该模块又把拉上来的cpu给offline了。
不过如此修改代码后，需要手动降低性能：
#offline A57
echo 0 > /sys/devices/system/cpu/cpu4/online
echo 0 > /sys/devices/system/cpu/cpu5/online
echo 0 > /sys/devices/system/cpu/cpu6/online
echo 0 > /sys/devices/system/cpu/cpu7/online
# down max freq
echo 960000 > /sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq
echo simple_ondemand > /sys/class/kgsl/kgsl-3d0/devfreq/governor
echo 5 > /sys/class/kgsl/kgsl-3d0/min_pwrlevel