TX2 PMC reset source: Denvor watchdog time out

Autonomous Machines Jetson & Embedded Systems Jetson TX2
1

Hi Guys,

When run our deep-learning application, sometimes TX2 will reboot directly. We catch the dmesg of reboot. There are may be some clue that “PMC reset source: Denvor watchdog time out”. Can you help us figure out the root cause?

The attachment is whole dmesg info, and also highlight useful dmesg info below. Hope it is helpful.

[Fri Feb  5 12:50:12 2021] iommu: Adding device 3550000.xudc to group 23
[Fri Feb  5 12:50:12 2021] tegra-pmc c360000.pmc: scratch reg offset dts data not present
[Fri Feb  5 12:50:12 2021] tegra-pmc: ### PMC reset source: Denvor watchdog time out
[Fri Feb  5 12:50:12 2021] tegra-pmc: ### PMC reset level: L1
[Fri Feb  5 12:50:12 2021] tegra-pmc: ### PMC reset status reg: 0x9
[Fri Feb  5 12:50:12 2021] padctrl padctrl.0: Pad control driver tegra-pmc-padctrl registered
[Fri Feb  5 12:50:12 2021] tegra-pmc c360000.pmc: IO padctrl driver initialized
[Fri Feb  5 12:50:12 2021] tegra186-aowake c370000.pmc: WAKE_AOWAKE_CTRL_0 = 3
[Fri Feb  5 12:50:12 2021] tegra186-aowake c370000.pmc: WAKE_AOWAKE_CNTRL_24(PMU_INT) = 320
[Fri Feb  5 12:50:12 2021] iommu: Adding device 13e10000.host1x to group 24
[Fri Feb  5 12:50:12 2021] iommu: Adding device 13e10000.host1x:ctx0 to group 25

dmesg.pmic (69.7 KB)

3

hello 53216142,

may I know which JetPack release you’re working with.
while you’re running deep-learning application, how long does the shutdown issue happened?
please also enable tegrastats utility to monitor the memory and processor usage for reference,
thanks

4

HI JerryChang,

We use JetPack3.3.3. This issue happened random. Sometime happen within quick time, sometimes never happen with whole one day.
The tegratestas monitor info is attached.
tegra_stats.log (546.4 KB)

7

There is no update from you for a period, assuming this is not an issue any more.
Hence we are closing this topic. If need further support, please open a new one.
Thanks

hello 53216142,

are you able to narrow down the issue,
for example, please try to minimum the complexity of application level and reproduce the failure.

FYI,
we don’t have similar issue for random shutdown on reference platform with JetPack-3.3.3,
thanks