As I mentioned before that we don’t suggest modifying the configuration for OC event, please create custom power mode configuration instead.
MAXN is an unconstraint mode rather than the best performance mode.
You need to create the custom power mode based on the power consumption in your case. You can simply create it through power estimator tool
Hi, I add own config by powerestimator:
< POWER_MODEL ID=4 NAME=MODE_CHCNAV_W >
CPU_ONLINE CORE_0 1
CPU_ONLINE CORE_1 1
CPU_ONLINE CORE_2 1
CPU_ONLINE CORE_3 1
CPU_ONLINE CORE_4 1
CPU_ONLINE CORE_5 1
CPU_ONLINE CORE_6 1
CPU_ONLINE CORE_7 1
FBP_POWER_GATING FBP_PG_MASK 2
TPC_POWER_GATING TPC_PG_MASK 240
GPU_POWER_CONTROL_ENABLE GPU_PWR_CNTL_EN on
CPU_A78_0 MIN_FREQ 729600
CPU_A78_0 MAX_FREQ 1984000
CPU_A78_1 MIN_FREQ 729600
CPU_A78_1 MAX_FREQ 1984000
GPU MIN_FREQ 0
GPU MAX_FREQ 918000000
GPU_POWER_CONTROL_DISABLE GPU_PWR_CNTL_DIS auto
EMC MAX_FREQ 3199000000
DLA0_CORE MAX_FREQ 614400000
DLA1_CORE MAX_FREQ 614400000
DLA0_FALCON MAX_FREQ 294400000
DLA1_FALCON MAX_FREQ 294400000
PVA0_VPS MAX_FREQ 512000000
PVA0_AXI MAX_FREQ 35840000
Here is config file:
nvpmodel_24.1.conf.txt (5.9 KB)
Query power mode:
root@tegra-ubuntu:/home/nvidia/stress# nvpmodel -q
NV Power Mode: MODE_CHCNAV_W
4
Reboot device and the run stress command:
jetson_clocks
echo performance | tee /sys/devices/system/cpu/cpu*/cpufreq/scaling_governor
./matrixMulCUBLAS --sizemult=8 &
Check tegrastats:
01-27-2026 11:57:26 RAM 2550/15504MB (lfb 2865x4MB) SWAP 2/7752MB (cached 0MB) CPU [1%@1984,5%@255,3%@1984,0%@490,1%@1984,0%@358,0%@1984,100%@1984] EMC_FREQ 37%@3199 GR3D_FREQ 99%@[917,0] VIC_FREQ 729 APE 174 CV0@79.781C CPU@84.406C SOC2@84.218C SOC0@74.843C CV1@75.968C GPU@98.156C tj@98.156C SOC1@76.468C CV2@75.656C VDD_IN 24028mW/24182mW VDD_CPU_GPU_CV 13626mW/13735mW VDD_SOC 4358mW/4389mW
01-27-2026 11:57:27 RAM 2550/15504MB (lfb 2865x4MB) SWAP 2/7752MB (cached 0MB) CPU [1%@1984,5%@378,1%@1984,1%@1984,0%@254,3%@1984,0%@1984,100%@869] EMC_FREQ 37%@3199 GR3D_FREQ 99%@[917,0] VIC_FREQ 729 APE 174 CV0@79.843C CPU@84.187C SOC2@84.031C SOC0@74.968C CV1@75.968C GPU@98.093C tj@98.093C SOC1@76.562C CV2@75.718C VDD_IN 24066mW/24182mW VDD_CPU_GPU_CV 13626mW/13734mW VDD_SOC 4395mW/4389mW
01-27-2026 11:57:28 RAM 2550/15504MB (lfb 2865x4MB) SWAP 2/7752MB (cached 0MB) CPU [0%@259,2%@1808,1%@246,1%@1984,0%@1984,2%@325,0%@1479,100%@1984] EMC_FREQ 37%@3199 GR3D_FREQ 99%@[918,0] VIC_FREQ 729 APE 174 CV0@79.812C CPU@84.5C SOC2@84.218C SOC0@74.843C CV1@76.156C GPU@98.375C tj@98.375C SOC1@76.406C CV2@75.687C VDD_IN 24178mW/24182mW VDD_CPU_GPU_CV 13737mW/13734mW VDD_SOC 4395mW/4389mW
01-27-2026 11:57:30 RAM 2551/15504MB (lfb 2865x4MB) SWAP 2/7752MB (cached 0MB) CPU [3%@250,4%@1984,4%@1984,1%@1250,0%@2133,0%@850,0%@1984,100%@523] EMC_FREQ 37%@3199 GR3D_FREQ 99%@[918,0] VIC_FREQ 729 APE 174 CV0@79.875C CPU@84.25C SOC2@84.218C SOC0@74.906C CV1@75.906C GPU@98.343C tj@98.343C SOC1@76.531C CV2@75.718C VDD_IN 24103mW/24182mW VDD_CPU_GPU_CV 13700mW/13734mW VDD_SOC 4395mW/4389mW
01-27-2026 11:57:31 RAM 2551/15504MB (lfb 2865x4MB) SWAP 2/7752MB (cached 0MB) CPU [3%@305,2%@1984,2%@1984,2%@490,1%@1582,1%@1984,0%@898,100%@1984] EMC_FREQ 37%@3199 GR3D_FREQ 99%@[917,0] VIC_FREQ 729 APE 174 CV0@79.937C CPU@84.25C SOC2@84.218C SOC0@74.75C CV1@76.031C GPU@98.312C tj@98.312C SOC1@76.593C CV2@75.812C VDD_IN 24140mW/24182mW VDD_CPU_GPU_CV 13663mW/13734mW VDD_SOC 4395mW/4389mW
01-27-2026 11:57:32 RAM 2551/15504MB (lfb 2865x4MB) SWAP 2/7752MB (cached 0MB) CPU [2%@251,1%@1831,4%@348,1%@1984,0%@409,3%@371,0%@1984,100%@322] EMC_FREQ 37%@3199 GR3D_FREQ 99%@[917,0] VIC_FREQ 729 APE 174 CV0@79.781C CPU@84.25C SOC2@84.187C SOC0@74.875C CV1@76C GPU@98.437C tj@98.437C SOC1@76.406C CV2@75.687C VDD_IN 23991mW/24181mW VDD_CPU_GPU_CV 13589mW/13734mW VDD_SOC 4358mW/4389mW
01-27-2026 11:57:33 RAM 2550/15504MB (lfb 2865x4MB) SWAP 2/7752MB (cached 0MB) CPU [0%@1984,0%@1984,5%@1055,2%@2153,2%@1984,2%@247,0%@1984,100%@250] EMC_FREQ 37%@3199 GR3D_FREQ 99%@[918,0] VIC_FREQ 729 APE 174 CV0@79.781C CPU@84.187C SOC2@84.218C SOC0@74.875C CV1@76C GPU@98.468C tj@98.468C SOC1@76.5C CV2@75.75C VDD_IN 24140mW/24181mW VDD_CPU_GPU_CV 13700mW/13734mW VDD_SOC 4395mW/4389mW
01-27-2026 11:57:34 RAM 2549/15504MB (lfb 2865x4MB) SWAP 2/7752MB (cached 0MB) CPU [2%@299,1%@1013,3%@1984,0%@1984,0%@1984,5%@1984,0%@1984,100%@1984] EMC_FREQ 37%@3199 GR3D_FREQ 99%@[917,0] VIC_FREQ 729 APE 174 CV0@79.781C CPU@84.25C SOC2@84.187C SOC0@74.937C CV1@75.968C GPU@98.593C tj@98.593C SOC1@76.562C CV2@75.687C VDD_IN 24178mW/24181mW VDD_CPU_GPU_CV 13737mW/13734mW VDD_SOC 4395mW/4389mW
01-27-2026 11:57:35 RAM 2549/15504MB (lfb 2865x4MB) SWAP 2/7752MB (cached 0MB) CPU [3%@1984,1%@1984,2%@372,1%@250,0%@1984,2%@1984,1%@1984,100%@1984] EMC_FREQ 37%@3199 GR3D_FREQ 99%@[918,0] VIC_FREQ 729 APE 174 CV0@79.906C CPU@84.218C SOC2@84.156C SOC0@74.906C CV1@76C GPU@98.375C tj@98.375C SOC1@76.562C CV2@75.781C VDD_IN 24140mW/24181mW VDD_CPU_GPU_CV 13700mW/13734mW VDD_SOC 4395mW/4389mW
01-27-2026 11:57:36 RAM 2550/15504MB (lfb 2865x4MB) SWAP 2/7752MB (cached 0MB) CPU [0%@1984,2%@1984,4%@324,1%@1776,1%@1984,0%@1408,1%@362,100%@1984] EMC_FREQ 37%@3199 GR3D_FREQ 99%@[917,0] VIC_FREQ 729 APE 174 CV0@79.937C CPU@84.156C SOC2@84.218C SOC0@74.843C CV1@76.031C GPU@98.156C tj@98.156C SOC1@76.343C CV2@75.812C VDD_IN 24066mW/24181mW VDD_CPU_GPU_CV 13589mW/13733mW VDD_SOC 4395mW/4389mW
01-27-2026 11:57:37 RAM 2549/15504MB (lfb 2865x4MB) SWAP 2/7752MB (cached 0MB) CPU [1%@1984,2%@1984,5%@1835,1%@1724,0%@1984,2%@1984,0%@255,100%@310] EMC_FREQ 37%@3199 GR3D_FREQ 99%@[918,0] VIC_FREQ 729 APE 174 CV0@79.843C CPU@84.406C SOC2@84.218C SOC0@74.906C CV1@76.156C GPU@98.312C tj@98.312C SOC1@76.375C CV2@75.75C VDD_IN 24178mW/24181mW VDD_CPU_GPU_CV 13737mW/13733mW VDD_SOC 4395mW/4389mW
01-27-2026 11:57:38 RAM 2549/15504MB (lfb 2865x4MB) SWAP 2/7752MB (cached 0MB) CPU [1%@1984,2%@1984,4%@1984,4%@297,1%@1984,2%@1984,0%@1557,100%@745] EMC_FREQ 37%@3199 GR3D_FREQ 99%@[918,0] VIC_FREQ 729 APE 174 CV0@79.781C CPU@84.562C SOC2@84.25C SOC0@74.843C CV1@75.968C GPU@98.281C tj@98.281C SOC1@76.5C CV2@75.687C VDD_IN 24028mW/24180mW VDD_CPU_GPU_CV 13589mW/13733mW VDD_SOC 4395mW/4389mW
01-27-2026 11:57:39 RAM 2548/15504MB (lfb 2865x4MB) SWAP 2/7752MB (cached 0MB) CPU [3%@1984,1%@1984,4%@354,0%@1984,2%@1984,0%@1984,2%@259,100%@340] EMC_FREQ 37%@3199 GR3D_FREQ 99%@[917,0] VIC_FREQ 729 APE 174 CV0@79.781C CPU@84.312C SOC2@84.218C SOC0@74.875C CV1@75.906C GPU@98.218C tj@98.218C SOC1@76.5C CV2@75.718C VDD_IN 24028mW/24180mW VDD_CPU_GPU_CV 13589mW/13733mW VDD_SOC 4358mW/4389mW
root@tegra-ubuntu:/home/nvidia/stress# grep "" /sys/class/hwmon/hwmon*/oc*_event_cnt
/sys/class/hwmon/hwmon1/oc1_event_cnt:920396
/sys/class/hwmon/hwmon1/oc2_event_cnt:73607
/sys/class/hwmon/hwmon1/oc3_event_cnt:250829
When setting up nvpmodel.conf for Orin NX 16G, I ensured it was within 25W. Why does running only the GPU stress test program still trigger a frequency drop?
As the CPU usage is low during stress test, could you reduce them to check if it could help for the OC issue?
An instantaneous power surge may exceed the OC (Overcurrent) threshold, triggering an OC event.
Hi,KevinFFF:
I downloaded a stress testing program for GPU on GitHub, and even if the GPU usage is set to only 50%, it will trigger an OC event.
Have you tried to reduce the CPU frequency in power mode configuration?
Hi,KevinFFF:
I tried setting the power mode to 25W:
root@tegra-ubuntu:/home/nvidia/stress# nvpmodel -q
NV Power Mode: 25W
3
Running in 25W mode will not trigger OC event.
Then I modified the configuration to only reduce the CPU frequency(1497MHz) before conducting further testing:
< POWER_MODEL ID=4 NAME=MODE_CHCNAV_W >
CPU_ONLINE CORE_0 1
CPU_ONLINE CORE_1 1
CPU_ONLINE CORE_2 1
CPU_ONLINE CORE_3 1
CPU_ONLINE CORE_4 1
CPU_ONLINE CORE_5 1
CPU_ONLINE CORE_6 1
CPU_ONLINE CORE_7 1
FBP_POWER_GATING FBP_PG_MASK 2
TPC_POWER_GATING TPC_PG_MASK 240
GPU_POWER_CONTROL_ENABLE GPU_PWR_CNTL_EN on
CPU_A78_0 MIN_FREQ 729600
CPU_A78_0 MAX_FREQ 1497600
CPU_A78_1 MIN_FREQ 729600
CPU_A78_1 MAX_FREQ 1497600
GPU MIN_FREQ 0
GPU MAX_FREQ 918000000
GPU_POWER_CONTROL_DISABLE GPU_PWR_CNTL_DIS auto
EMC MAX_FREQ 3199000000
DLA0_CORE MAX_FREQ 614400000
DLA1_CORE MAX_FREQ 614400000
DLA0_FALCON MAX_FREQ 294400000
DLA1_FALCON MAX_FREQ 294400000
PVA0_VPS MAX_FREQ 512000000
Perform stress testing, there was still a decrease in frequency:
Check the power supply, it has exceeded 25W.
Hi,KevinFFF:
Previously, I used an external fan to blow at the device without reducing the frequency. At that time, I was using JetPack 5.1.3. Now, I have flashed JetPack 5.1.5 (which is the version we have confirmed to use), which immediately triggers the OC event regardless of the environment test. What is the difference between these two versions?
Good to hear that the using the different power mode configuration may help for your case.
Maybe you can try to disable few CPU cores as they seem not needed in your test case.
You can compare if the power mode configuration is different in these 2 releases.
As the Jetpack 5.1.6 has been released this week, you can also give it a try to check if it helps in your case.
In 25W mode, neither CPU nor GPU has the maximum frequency. The CPU is set to a frequency of 1.5G and 612M.
My requirement is that all 8 CPU cores are running, and the algorithm program that needs to be run requires Orin NX to run at full power.
We have confirmed the use of JetPack5.1.5, which is a requirement of the algorithm. Now I just want to find the differences between JetPack5.1.3 and JetPack5.1.5 cores, and understand how to modify them to avoid frequency reduction like 5.1.3。
If you don’t disable few cores(even with low loading on them), you might hit OC event issue and get throttling. To get better performance, we would suggest creating the custom power mode configuration for specific use case.
Do you mean there’s the issue in Jetpack 5.1.5 but not for Jetpack5.1.3 so you want to find the difference?
If so, please provide the result of the following command in these 2 cases.
# nvpmodel -q --verbose
Hi,KevinFFF:
JetPack5.1.5:
Jetpack 5.1.5 [L4T 35.6.1]：
root@tegra-ubuntu:/home/nvidia# nvpmodel -q --verbose
NVPM VERB: Config file: /etc/nvpmodel.conf
NVPM VERB: parsing done for /etc/nvpmodel.conf
NVPM VERB: Current mode: NV Power Mode: MAXN
0
NVPM VERB: PARAM CPU_ONLINE: ARG CORE_0: PATH /sys/devices/system/cpu/cpu0/online: REAL_VAL: 1 CONF_VAL: 1
NVPM VERB: PARAM CPU_ONLINE: ARG CORE_1: PATH /sys/devices/system/cpu/cpu1/online: REAL_VAL: 1 CONF_VAL: 1
NVPM VERB: PARAM CPU_ONLINE: ARG CORE_2: PATH /sys/devices/system/cpu/cpu2/online: REAL_VAL: 1 CONF_VAL: 1
NVPM VERB: PARAM CPU_ONLINE: ARG CORE_3: PATH /sys/devices/system/cpu/cpu3/online: REAL_VAL: 1 CONF_VAL: 1
NVPM VERB: PARAM CPU_ONLINE: ARG CORE_4: PATH /sys/devices/system/cpu/cpu4/online: REAL_VAL: 1 CONF_VAL: 1
NVPM VERB: PARAM CPU_ONLINE: ARG CORE_5: PATH /sys/devices/system/cpu/cpu5/online: REAL_VAL: 1 CONF_VAL: 1
NVPM VERB: PARAM CPU_ONLINE: ARG CORE_6: PATH /sys/devices/system/cpu/cpu6/online: REAL_VAL: 1 CONF_VAL: 1
NVPM VERB: PARAM CP
NVPM VERB: PARAM FBP_POWER_GATING: ARG FBP_PG_MASK: PATH /sys/devices/gpu.0/fbp_pg_mask: REAL_VAL: 2 CONF_VAL: 2
NVPM VERB: PARAM TPC_POWER_GATING: ARG TPC_PG_MASK: PATH /sys/devices/gpu.0/tpc_pg_mask: REAL_VAL: 240 CONF_VAL: 240
NVPM VERB: PARAM GPU_POWER_CONTROL_ENABLE: ARG GPU_PWR_CNTL_EN: PATH /sys/devices/gpu.0/power/control: REAL_VAL: auto CONF_VAL: on
NVPM VERB: PARAM CPU_A78_0: ARG MIN_FREQ: PATH /sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq: REAL_VAL: 1984000 CONF_VAL: 729600
NVPM VERB: PARAM CPU_A78_0: ARG MAX_FREQ: PATH /sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq: REAL_VAL: 1984000 CONF_VAL: 9223372036854775807
NVPM VERB: PARAM CPU_A78_1: ARG MIN_FREQ: PATH /sys/devices/system/cpu/cpu4/cpufreq/scaling_min_freq: REAL_VAL: 1984000 CONF_VAL: 729600
NVPM VERB: PARAM CPU_A78_1: ARG MAX_FREQ: PATH /sys/devices/system/cpu/cpu4/cpufreq/scaling_max_freq: REAL_VAL: 1984000 CONF_VAL: 9223372036854775807
NVPM VERB: PARAM GPU: ARG MIN_FREQ: PATH /sys/devices/1700
NVPM VERB: PARAM DLA0_CORE: ARG MAX_FREQ: PATH /sys/devices/platform/13e40000.host1x/15880000.nvdla0/acm/clk_cap/dla0_core: REAL_VAL: 614400000 CONF_VAL: 9223372036854775807
NVPM VERB: PARAM DLA1_CORE: ARG MAX_FREQ: PATH /sys/devices/platform/13e40000.host1x/158c0000.nvdla1/acm/clk_cap/dla1_core: REAL_VAL: 614400000 CONF_VAL: 9223372036854775807
NVPM VERB: PARAM DLA0_FALCON: ARG MAX_FREQ: PATH /sys/devices/platform/13e40000.host1x/15880000.nvdla0/acm/clk_cap/dla0_falcon: REAL_VAL: 294400000 CONF_VAL: 9223372036854775807
NVPM VERB: PARAM DLA1_FALCON: ARG MAX_FREQ: PATH /sys/devices/platform/13e40000.host1x/158c0000.nvdla1/acm/clk_cap/dla1_falcon: REAL_VAL: 294400000 CONF_VAL: 9223372036854775807
NVPM VERB: PARAM PVA0_VPS: ARG MAX_FREQ: PATH /sys/devices/platform/13e40000.host1x/16000000.pva0/acm/clk_cap/pva0_vps: REAL_VAL: 1190400000 CONF_VAL: 9223372036854775807
NVPM VERB: PARAM PVA0_AXI: ARG MAX_FREQ: PATH /sys/devices/platform/13e40000.host1x/16000000.pva0/acm/clk_cap/pva0_cpu_a
JetPack5.1.3:
root@tegra-ubuntu:/home/nvidia# jetson_release
Software part of jetson-stats 4.3.2 - (c) 2024, Raffaello Bonghi
Model: NVIDIA Orin NX Developer Kit - Jetpack 5.1.3 [L4T 35.5.0]
NV Power Mode[0]: MAXN
Serial Number: [XXX Show with: jetson_release -s XXX]
Hardware:
- P-Number: p3767-0000
- Module: NVIDIA Jetson Orin NX (16GB ram)
Platform:
- Distribution: Ubuntu 20.04 focal
- Release: 5.10.192-tegra
jtop:
- Version: 4.3.2
- Service: Active
Libraries:
- CUDA: 11.4.315
- cuDNN: 8.6.0.166
- TensorRT: 8.5.2.2
- VPI: 2.4.8
- OpenCV: 4.5.4 - with CUDA: NO
root@tegra-ubuntu:/home/nvidia# nvpmodel -q --verbose
NVPM VERB: Config file: /etc/nvpmodel.conf
NVPM VERB: parsing done for /etc/nvpmodel.conf
NVPM VERB: Current mode: NV Power Mode: MAXN
0
NVPM VERB: PARAM CPU_ONLINE: ARG CORE_0: PATH /sys/devices/system/cpu/cpu0/online: REAL_VAL: 1 CONF_VAL: 1
NVPM VERB: PARAM CPU_ONLINE: ARG CORE_1: PATH /sys/devices/system/cpu/cpu1/online: REAL_VAL: 1 CONF_VAL: 1
NVPM VERB: PARAM CPU_ONLINE: ARG CORE_2: PATH /sys/devices/system/cpu/cpu2/online: REAL_VAL: 1 CONF_VAL: 1
NVPM VERB: PARAM CPU_ONLINE: ARG CORE_3: PATH /sys/devices/system/cpu/cpu3/online: REAL_VAL: 1 CONF_VAL: 1
NVPM VERB: PARAM CPU_ONLINE: ARG CORE_4: PATH /sys/devices/system/cpu/cpu4/online: REAL_VAL: 1 CONF_VAL: 1
NVPM VERB: PARAM CPU_ONLINE: ARG CORE_5: PATH /sys/devices/system/cpu/cpu5/online: REAL_VAL: 1 CONF_VAL: 1
NVPM VERB: PARAM CPU_ONLINE: ARG CORE_6: PATH /sys/devices/system/cpu/cpu6/online: REAL_VAL: 1 CONF_VAL: 1
NVPM VERB: PARAM CPU_ONLINE: ARG CORE_7: PATH /sys/devices/system/cpu/cpu7/online: REAL_VAL: 1 CONF_VAL: 1
NVPM VERB: PARAM FBP_POWER_GATING: ARG FBP_PG_MASK: PATH /sys/devices/gpu.0/fbp_pg_mask: REAL_VAL: 2 CONF_VAL: 2
NVPM VERB: PARAM TPC_POWER_GATING: ARG TPC_PG_MASK: PATH /sys/devices/gpu.0/tpc_pg_mask: REAL_VAL: 240 CONF_VAL: 240
NVPM VERB: PARAM GPU_POWER_CONTROL_ENABLE: ARG GPU_PWR_CNTL_EN: PATH /sys/devices/gpu.0/power/control: REAL_VAL: auto CONF_VAL: on
NVPM VERB: PARAM CPU_A78_0: ARG MIN_FREQ: PATH /sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq: REAL_VAL: 1984000 CONF_VAL: 729600
NVPM VERB: PARAM CPU_A78_0: ARG MAX_FREQ: PATH /sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq: REAL_VAL: 1984000 CONF_VAL: 9223372036854775807
NVPM VERB: PARAM CPU_A78_1: ARG MIN_FREQ: PATH /sys/devices/system/cpu/cpu4/cpufreq/scaling_min_freq: REAL_VAL: 1984000 CONF_VAL: 729600
NVPM VERB: PARAM CPU_A78_1: ARG MAX_FREQ: PATH /sys/devices/system/cpu/cpu4/cpufreq/scaling_max_freq: REAL_VAL: 1984000 CONF_VAL: 9223372036854775807
NVPM VERB: PARAM GPU: ARG MIN_FREQ: PATH /sys/devices/17000000.ga10b/devfreq_dev/min_freq: REAL_VAL: 918000000 CONF_VAL: 0
NVPM VERB: PARAM GPU: ARG MAX_FREQ: PATH /sys/devices/17000000.ga10b/devfreq_dev/max_freq: REAL_VAL: 918000000 CONF_VAL: 9223372036854775807
NVPM VERB: PARAM GPU_POWER_CONTROL_DISABLE: ARG GPU_PWR_CNTL_DIS: PATH /sys/devices/gpu.0/power/control: REAL_VAL: auto CONF_VAL: auto
NVPM VERB: PARAM EMC: ARG MAX_FREQ: PATH /sys/kernel/nvpmodel_emc_cap/emc_iso_cap: REAL_VAL: 0 CONF_VAL: 0
NVPM VERB: PARAM DLA0_CORE: ARG MAX_FREQ: PATH /sys/devices/platform/13e40000.host1x/15880000.nvdla0/acm/clk_cap/dla0_core: REAL_VAL: 614400000 CONF_VAL: 9223372036854775807
NVPM VERB: PARAM DLA1_CORE: ARG MAX_FREQ: PATH /sys/devices/platform/13e40000.host1x/158c0000.nvdla1/acm/clk_cap/dla1_core: REAL_VAL: 614400000 CONF_VAL: 9223372036854775807
NVPM VERB: PARAM DLA0_FALCON: ARG MAX_FREQ: PATH /sys/devices/platform/13e40000.host1x/15880000.nvdla0/acm/clk_cap/dla0_falcon: REAL_VAL: 294400000 CONF_VAL: 9223372036854775807
NVPM VERB: PARAM DLA1_FALCON: ARG MAX_FREQ: PATH /sys/devices/platform/13e40000.host1x/158c0000.nvdla1/acm/clk_cap/dla1_falcon: REAL_VAL: 294400000 CONF_VAL: 9223372036854775807
NVPM VERB: PARAM PVA0_VPS: ARG MAX_FREQ: PATH /sys/devices/platform/13e40000.host1x/16000000.pva0/acm/clk_cap/pva0_vps: REAL_VAL: 704000000 CONF_VAL: 9223372036854775807
NVPM VERB: PARAM PVA0_AXI: ARG MAX_FREQ: PATH /sys/devices/platform/13e40000.host1x/16000000.pva0/acm/clk_cap/pva0_cpu_axi: REAL_VAL: 486400000 CONF_VAL: 9223372036854775807
After comparing the two versions, the frequency setting of PVA is different:
It’s a good finding.
Could you try configuring 704MHz for pva0_vps and 486.4MHz for pva0_cpu_axi in Jetpack 5.1.5 to check if it can help for your case?