summaryrefslogtreecommitdiff
path: root/tools/aapt2/java/JavaClassGenerator_test.cpp
diff options
context:
space:
mode:
authorHui Yu <huiyu@google.com>2018-05-23 14:39:24 -0700
committerHui Yu <huiyu@google.com>2018-05-24 00:45:02 +0000
commit39b29bc2303c2bb346cba705c8903062af3ff031 (patch)
tree1d7f690b448a4ea55f4fe32a322e1e99986b6a79 /tools/aapt2/java/JavaClassGenerator_test.cpp
parent57f6932f29bfed9d42743c1a99a47779184dc7c5 (diff)
Calculate cellular radio value from modem.controller values.
Recently cellular radio related values {"radio.active", "radio.scanning", "radio.on"} are removed from power_profile.xml, which causes inaccurate radio power usage. We use following formula to calculate radio values from modem.controller values. radio.active = average of modem.controller.rx and modem.controller.tx values radio.scanning = 0 (since this is already included in radio.on bin 0) <array name="radio.on"> <!-- Strength 0 to BINS-1 --> <value>modem.controller.idle * 25 / 180 </value> <!-- none --> <value>max(1, modem.controller.idle/256) </value> <!-- poor --> <value>max(1, modem.controller.idle/256) </value> <!-- moderate --> <value>max(1, modem.controller.idle/256) </value> <!-- good --> <value>max(1, modem.controller.idle/256) </value> <!-- great --> </array> Bug: 79379255 Test: use debugger to observe correct mPowerRadioOn, mPowerBins, mPowerScan are calculated. "adb shell dumpsys batterystats", looking for "radio=" in "Estimated power use" section. Change-Id: Ic65a5c8a35a5b1f4ba05ddc150e29b00bc62689f
Diffstat (limited to 'tools/aapt2/java/JavaClassGenerator_test.cpp')
0 files changed, 0 insertions, 0 deletions