Thermal loop doesn't distinguish between "real" and "fake" temperatures #2385
Description
➜ ~ humility -d 1001-sled26_hubris.core.0 ringbuf thermal
humility: attached to dump
humility: ring buffer drv_i2c_devices::emc2305::__RINGBUF in thermal:
humility: ring buffer drv_i2c_devices::max31790::__RINGBUF in thermal:
humility: ring buffer task_thermal::__RINGBUF in thermal:
TOTAL VARIANT
7545 ControlPwm
57 SensorReadFailed
3 AutoState(Boot)
2 AutoState(Running)
1 AutoState(Overheated)
1 AutoState(Uncontrollable)
6 FanAdded
3 MiscReadFailed
2 PowerModeChanged
1 Start
1 ThermalMode(Auto)
1 PowerDownDueTo
1 CriticalDueTo
1 FanControllerInitialized
1 PowerDownAt
1 SetFanWatchdogOk
NDX LINE GEN COUNT PAYLOAD
30 895 4 1 SensorReadFailed(SensorId(0x38), I2cError(NoDevice))
31 1206 4 1 ControlPwm(0x57)
0 895 5 1 SensorReadFailed(SensorId(0x38), I2cError(NoDevice))
1 1112 5 1 CriticalDueTo { sensor_id: SensorId(0x38), temperature: Celsius(80.3285) }
2 1120 5 1 AutoState(Overheated)
3 1206 5 1 ControlPwm(0x64)
4 895 5 1 SensorReadFailed(SensorId(0x38), I2cError(NoDevice))
5 1206 5 1 ControlPwm(0x64)
6 895 5 1 SensorReadFailed(SensorId(0x38), I2cError(NoDevice))
7 1206 5 1 ControlPwm(0x64)
8 895 5 1 SensorReadFailed(SensorId(0x38), I2cError(NoDevice))
9 1206 5 1 ControlPwm(0x64)
10 895 5 1 SensorReadFailed(SensorId(0x38), I2cError(NoDevice))
11 1206 5 1 ControlPwm(0x64)
12 895 5 1 SensorReadFailed(SensorId(0x38), I2cError(NoDevice))
13 1206 5 1 ControlPwm(0x64)
14 895 5 1 SensorReadFailed(SensorId(0x38), I2cError(NoDevice))
15 1206 5 1 ControlPwm(0x64)
16 895 5 1 SensorReadFailed(SensorId(0x38), I2cError(NoDevice))
17 1206 5 1 ControlPwm(0x64)
18 895 5 1 SensorReadFailed(SensorId(0x38), I2cError(NoDevice))
19 1206 5 1 ControlPwm(0x64)
20 895 5 1 SensorReadFailed(SensorId(0x38), I2cError(NoDevice))
21 1206 5 1 ControlPwm(0x64)
22 895 5 1 SensorReadFailed(SensorId(0x38), I2cError(NoDevice))
23 1164 5 1 PowerDownDueTo { sensor_id: SensorId(0x38), temperature: Celsius(85.351) }
24 1169 5 1 AutoState(Uncontrollable)
25 1210 5 1 PowerDownAt(0x72435)
26 964 5 1 PowerModeChanged(PowerBitmask(0b1))
27 788 5 1 AutoState(Boot)
28 1065 5 1 AutoState(Running)
29 1206 5 7085 ControlPwm(0x0)
While debugging #2384, we noticed that the temperatures reported here may actually be artificially inflated due to this logic
but we don't indicate this or otherwise provide any indication that this might not actually be a real, measured temperature, and as such this can be extremely confusing. We should provide some way of indicating that this is a synthesized temperature in this case.