Hi NVidia Team
We created our own power mode, where the only difference to the MAXN mode is that the TCP_POWER_GATING mask is 5 and not 0 as in MAXN. In this mode our application runs fine and does not crash.
If we use the MAXN mode and we start our application, it crashes after 3 seconds. Temperatures do not exceed 50°C. Has the TCP its own voltage rail and clock speed? What could be the problem?
Thank you.
Kind regards
hello sevm89,
according to NVPModel Clock Configuration for Jetson AGX Xavier, you should note that MaxN did not define the power budget.
could you please enable tegrastats utility and monitor the difference.
thanks
Hi JerryChang
We are aware that MAXN did not define the power budget. Still we would like to know why changing the paramter TCP_POWER_GATING from 5 to 0 leads to a crash. Before the crash we do not see significant differences between the output of the tegrastats utility, even though in MAXN we think that we can not see anything special in this 3 seconds.
Kind regars
Hi JerryChang
Any news on the problem? Has this parameter influence on the 5V current consumption or the 12V?
Here the output of tegrastats before the crash:
RAM 6735/15698MB (lfb 1726x4MB) SWAP 0/7849MB (cached 0MB) CPU [73%@2265,44%@2265,59%@2265,41%@2265,48%@2265,38%@2265,47%@2265,45%@2265] EMC_FREQ 0% GR3D_FREQ 99% AO@34C GPU@38C Tboard@34C Tdiode@36.25C AUX@33.5C CPU@38.5C thermal@36.35C PMIC@100C GPU 10326/9950 CPU 6314/6458 SOC 4688/4607 CV 0/0 VDDRQ 1817/1754 SYS5V 4753/4728
RAM 6748/15698MB (lfb 1722x4MB) SWAP 0/7849MB (cached 0MB) CPU [78%@2265,46%@2265,60%@2265,48%@2265,47%@2265,50%@2265,49%@2265,47%@2265] EMC_FREQ 0% GR3D_FREQ 99% AO@34C GPU@38.5C Tboard@34C Tdiode@36.25C AUX@33.5C CPU@38.5C thermal@36.35C PMIC@100C GPU 10613/9954 CPU 6788/6460 SOC 4688/4607 CV 0/0 VDDRQ 1817/1754 SYS5V 4753/4728
RAM 6758/15698MB (lfb 1719x4MB) SWAP 0/7849MB (cached 0MB) CPU [81%@2265,42%@2265,62%@2265,52%@2265,52%@2265,50%@2265,48%@2265,45%@2265] EMC_FREQ 0% GR3D_FREQ 99% AO@34C GPU@38C Tboard@34C Tdiode@36.5C AUX@33.5C CPU@38.5C thermal@36.05C PMIC@100C GPU 10422/9956 CPU 6693/6461 SOC 4688/4608 CV 0/0 VDDRQ 1817/1754 SYS5V 4753/4728
RAM 6767/15698MB (lfb 1717x4MB) SWAP 0/7849MB (cached 0MB) CPU [77%@2265,44%@2265,64%@2265,51%@2265,50%@2265,50%@2265,47%@2265,47%@2265] EMC_FREQ 0% GR3D_FREQ 99% AO@34C GPU@38C Tboard@34C Tdiode@36.25C AUX@33.5C CPU@38.5C thermal@36.05C PMIC@100C GPU 10613/9960 CPU 6697/6462 SOC 4688/4608 CV 0/0 VDDRQ 1817/1755 SYS5V 4753/4728
RAM 6777/15698MB (lfb 1715x4MB) SWAP 0/7849MB (cached 0MB) CPU [72%@2265,42%@2265,62%@2265,49%@2265,53%@2265,51%@2265,48%@2265,45%@2265] EMC_FREQ 0% GR3D_FREQ 99% AO@34C GPU@38C Tboard@34C Tdiode@36.25C AUX@33.5C CPU@38.5C thermal@36.2C PMIC@100C GPU 10517/9963 CPU 6693/6463 SOC 4685/4609 CV 0/0 VDDRQ 1817/1755 SYS5V 4753/4728
RAM 6787/15698MB (lfb 1713x4MB) SWAP 0/7849MB (cached 0MB) CPU [76%@2265,44%@2265,60%@2265,44%@2265,52%@2265,53%@2265,46%@2265,43%@2265] EMC_FREQ 0% GR3D_FREQ 99% AO@34.5C GPU@38C Tboard@34C Tdiode@36.25C AUX@33.5C CPU@38.5C thermal@36.35C PMIC@100C GPU 10517/9965 CPU 6601/6464 SOC 4688/4609 CV 0/0 VDDRQ 1817/1755 SYS5V 4753/4728
RAM 6791/15698MB (lfb 1711x4MB) SWAP 0/7849MB (cached 0MB) CPU [77%@2265,44%@2265,63%@2265,46%@2265,55%@2265,51%@2265,50%@2265,50%@2265] EMC_FREQ 0% GR3D_FREQ 99% AO@34.5C GPU@38C Tboard@34C Tdiode@36.5C AUX@33.5C CPU@38.5C thermal@36.5C PMIC@100C GPU 10135/9966 CPU 6793/6466 SOC 4688/4609 CV 0/0 VDDRQ 1819/1756 SYS5V 4753/4728
Thank you.
hello sevm89,
were there kernel failures which could indicate the crash in detail.
you may setup serial console to check the logs dynamically, or you could check it with $ dmesg --follow