Notes on Sony Xperia devices

[hellobbn]

Hi,

I am creating this issue to provide some information on the specific mechanism sony's own implementation of "battery idle mode" (Which sony calls it H.S. mode).

TL; DR To enable battery idle mode, issue the following command to the kernel interface:

# Activate smart charging
echo 1 > /sys/class/power_supply/battery_ext/smart_charging_activiation

# Enable battery idle mode
echo 1 > /sys/class/power_supply/battery_ext/smart_charging_interruption

To disable:

echo 0 > /sys/class/power_supply/battery_ext/smart_charging_interruption

Detail:

In stock sony devices, this battery idle mode can be enabled in settings, and when it is enabled, it makes a call to sony's charger management HAL, which will set the property persist.vendor.battery.smt_chg, and the init file will write to the kernel interface mentioned above when a property change is noticed:

on property:persist.vendor.battery.smt_chg=activate
    write /sys/class/power_supply/battery_ext/smart_charging_activation 1

on property:persist.vendor.battery.smt_chg=charging_suspend
    write /sys/class/power_supply/battery_ext/smart_charging_interruption 1

on property:persist.vendor.battery.smt_chg=charging_resume
    write /sys/class/power_supply/battery_ext/smart_charging_interruption 0

Is it possible for that to be implemented in acc?

[VR-25]

Sure. ACC supports multiple charging switches (e.g., acc --set charging_switch="sw1 on off sw2 on off...") New pre-release coming up soon.

[hellobbn]

I am sorry, I made a typo here, it is not smart_charging_activiation but smart_charging_activation

Sorry for the mistake

[hellobbn]

After the correction, the test does not seem to work properly:

6/24: battery_ext/smart_charging_activation 0 1 battery_ext/smart_charging_interruption 0 1
  switch: off (1, 1)    current: 1934mA (Charging)
  switch: off (1, 1)    current: -66mA (Discharging)
  switch: on (1)    current: -66mA (Discharging)
  Switch doesn't work ❌
  ##########
  Waiting for charging to resume...
  If it doesn't happen after a few seconds, try re-plugging the charger.
  If it's taking too long, unplug the charger, stop the test with CTRL-C, run accd -i, wait a few seconds, and retest.
  In extreme cases, one shall comment out (blacklist) this switch in /data/adb/vr25/acc-data/logs/write.log, reboot (to enable charging), and re-run the test.
  ##########
  switch: - (1) current: -66mA (Discharging)
  switch: - (1) current: 0mA (Idle)
  switch: - (1) current: 0mA (Idle)
  switch: - (1) current: -1mA (Idle)
  switch: - (1) current: -0mA (Idle)
  switch: - (1) current: -0mA (Idle)
  switch: - (1) current: -0mA (Idle)
  switch: - (1) current: -0mA (Idle)
  switch: - (1) current: -1mA (Idle)
  switch: - (1) current: -1mA (Idle)
  switch: - (1) current: -66mA (Discharging)
  switch: - (1) current: 0mA (Idle)
  switch: - (1) current: 0mA (Idle)
  switch: - (1) current: 0mA (Idle)
  switch: - (1) current: 0mA (Idle)
  switch: - (1) current: -1mA (Idle)
  switch: - (1) current: -1mA (Idle)
  switch: - (1) current: -1mA (Idle)
  switch: - (1) current: -0mA (Idle)
  switch: - (1) current: -0mA (Idle)
  switch: - (1) current: -0mA (Idle)

[hellobbn]

The problem is possibly because that the interface /sys/class/power_supply/battery_ext/smart_charging_activation cannot be set to 0 once it is 1 (And it is expected behavior).

[VR-25]

Ok, I see. Will push changes accordingly soon.

On Sat, Jul 30, 2022, 01:55 hellobbn @.***> wrote:

The problem is possibly because that the interface /sys/class/power_supply/battery_ext/smart_charging_activation cannot be set to 0 once it is 1 (And it is expected behavior).

— Reply to this email directly, view it on GitHub https://github.com/VR-25/acc/issues/163#issuecomment-1200053584, or unsubscribe https://github.com/notifications/unsubscribe-auth/AG42GSAMF3A3VV5HDVBHWD3VWR4QLANCNFSM54YBZMBQ . You are receiving this because you modified the open/close state.Message ID: @.***>

[hellobbn]

Thanks, really appreciate it.