-
Once critical overcurrent happen, just throttling? Recovered dynamically?
-
Is there FLAG not cleared automatically of overcurrent? I want to know overcurrent happened later.
hello kimjishu,
may I know what’s the actual use-case? this messages (i.e. throttling alarms) means it’s reaching the hardware spec limitation and it’s trying to protect the module.
you could execute tegrastats utility to monitor the memory usage and also processor usage. the clock frequency will drops when you see throttling alarms.
Actual Use-Case : Decoding and Analysis(Heatmap) for 4 Visions
I can see console “soctherm: OC ALARM 0x00000010”
If I command below, OC ALARM is disaapear
sudo sh -c “echo 5000 > /sys/bus/i2c/drivers/ina3221x/2-0040/iio:device0/crit_current_limit_0”
If I command below, Current is about 2000mA
“/sys/bus/i2c/drivers/ina3221x/2-0040/iio:device0/in_current0_input”
Maybe “crit_current_limit_0” is seeing Peak Current and “in_current0_input” is seeing Average Current. Right?
So, Is It problem? “soctherm: OC ALARM 0x00000010”
And what does It mean?
I have several Issues But looks like Software Issue yet…
hello kimjishu,
OC ALARM it means over-current events, which are handled in hardware.
when OC throttling is happening, there is an interrupt from bpmp to cpu, the tegra_oc_event driver detects the interrupt and then increments the interrupt count and prints those kernel logs.
you may also access TRM, please check SOC_THERM for more details.
this mean the thermal trip is enabled, and system will then be shutdown directly by PMIC instead of causing a shutdown through Boot ROM.
But shutdown doesn’t happen. When does shutdown happen?
Should I fix it “soctherm: OC ALARM 0x00000010” is printed?
hello kimjishu,
shutdown happen when it reaches the thermal trip point, you may also monitor the thermal by tegrastats.
and… you can modify crit_current_limit_0 to avoid this with your own product.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.