APE in for Audio processing engine setting

When I am using Xavier NX eMMC version to run the deepstream sample demo
I find out APE (audio processing engine) number is different when I am running the same model in Jetson stats. (also in Tegra and CLK documents)

The APE in one machine is 75Hmz while the other three are all 150 Mhz
I found a table that said Xavier NX is 150 Mhz on APE.

May I Know why it causes the difference between the two numbers in the same machine? Or is there a place or file I can do the setting on APE on myself?

H PeterYu,

Can you please confirm the JP version you are using Jetson NX?

I see that you are using JP4.4 Developer Preview image on the one showing 75MHz APE clock. Please check with latest JP4.4.1.

Thanks,
Sharad

Hi Sharadg

I have confirmed that
Both of the JP version are Jetapck 4.4 DP

Hi PeterYu,

There was a change done after (/during) JP4.4 DP release timeframe to increase the APE clock to 150MHz. Please use JP4.4.1. That should resolve the issue.

Thanks,
Sharad

Hi Sharadg

Do you mean in Jetpack 4.4 DP version,
The APE will appear 75MHz and 150 Mhz at the same time inappropriately?
I have 4 machines there, and 3 showing 150Mhz, and only 1 show 75Mhz.
I am sure that there are all Jetpack 4.4 DP version, I afraid it may not JP version issue that causes the difference

Could you please check “cat /sys/kernel/debug/bpmp/debug/clk/ape/max_rate” on these boards and paste outputs here?

Hi Sharadg

I got
75000000 (on the issue board)
and
150000000 respectivily (other 3)

Hi Peter,

We confirmed again at our end, max possible APE clock with JP4.4 DP release is 75MHz.

I assume that you are using Jetson NX devkit and not a custom HW. Assuming that,
75KHz is the right APE clock value if you are checking with JP4.4 DP release. Request you to move to JP 4.4.1 (latest release) if your usecase really needs 150MHz APE clock.

Thanks,
Sharad

https://docs.nvidia.com/jetson/l4t/index.html#page/Tegra%2520Linux%2520Driver%2520Package%2520Development%2520Guide%2Fpower_management_jetson_xavier.html%23wwpID0E0NO0HA

Above section explains various power modes on Jetson Xavier boards. Please check if there is a power mode difference between these which is causing the issue.

Also please share “dpkg -l | grep -i nvidia” output on the board showing 150MHz APE clock?

Thanks,
Sharad

Hi Sharadg

I have compared the 75 MHz one and 150MHz

their packages are as same as below

Blockquote
ii cuda-nvtx-10-2 10.2.89-1 arm64 NVIDIA Tools Extension
ii deepstream-5.0 5.0.0-1 arm64 Nvidia DeepStreamSDK runtime libraries, development files and samples
ii libnvidia-container-tools 0.9.0~beta.1 arm64 NVIDIA container runtime library (command-line tools)
ii libnvidia-container0:arm64 0.9.0~beta.1 arm64 NVIDIA container runtime library
ii libvisionworks 1.6.0.501 arm64 NVIDIA’s VisionWorks Library and supplemental data
ii libvisionworks-dev 1.6.0.501 all Development files for NVIDIA’s VisionWorks Library
ii libvisionworks-samples 1.6.0.501 arm64 Samples for NVIDIA’s VisionWorks Library
ii libvisionworks-sfm 0.90.4.501 arm64 SFM module for NVIDIA’s VisionWorks Library
ii libvisionworks-sfm-dev 0.90.4.501 arm64 Development files for SFM module for NVIDIA’s VisionWorks Library
ii libvisionworks-tracking 0.88.2.501 arm64 Tracking module for NVIDIA’s VisionWorks Library
ii libvisionworks-tracking-dev 0.88.2.501 arm64 Development files for Tracking module for NVIDIA’s VisionWorks Library
ii nvidia-container-csv-cuda 10.2.89-1 arm64 Jetpack CUDA CSV file
ii nvidia-container-csv-cudnn 8.0.0.145-1+cuda10.2 arm64 Jetpack CUDNN CSV file
ii nvidia-container-csv-tensorrt 7.1.0.16-1+cuda10.2 arm64 Jetpack TensorRT CSV file
ii nvidia-container-csv-visionworks 1.6.0.501 arm64 Jetpack VisionWorks CSV file
ii nvidia-container-runtime 3.1.0-1 arm64 NVIDIA container runtime
ii nvidia-container-toolkit 1.0.1-1 arm64 NVIDIA container runtime hook
ii nvidia-docker2 2.2.0-1 all nvidia-docker CLI wrapper
ii nvidia-l4t-3d-core 32.4.2-20200408182620 arm64 NVIDIA GL EGL Package
ii nvidia-l4t-apt-source 32.4.2-20200408182620 arm64 NVIDIA L4T apt source list debian package
ii nvidia-l4t-bootloader 32.4.2-20200408182620 arm64 NVIDIA Bootloader Package
ii nvidia-l4t-camera 32.4.2-20200408182620 arm64 NVIDIA Camera Package
ii nvidia-l4t-configs 32.4.2-20200408182620 arm64 NVIDIA configs debian package
ii nvidia-l4t-core 32.4.2-20200408182620 arm64 NVIDIA Core Package
ii nvidia-l4t-cuda 32.4.2-20200408182620 arm64 NVIDIA CUDA Package
ii nvidia-l4t-firmware 32.4.2-20200408182620 arm64 NVIDIA Firmware Package
ii nvidia-l4t-graphics-demos 32.4.2-20200408182620 arm64 NVIDIA graphics demo applications
ii nvidia-l4t-gstreamer 32.4.2-20200408182620 arm64 NVIDIA GST Application files
ii nvidia-l4t-init 32.4.2-20200408182620 arm64 NVIDIA Init debian package
ii nvidia-l4t-initrd 32.4.2-20200408182620 arm64 NVIDIA initrd debian package
ii nvidia-l4t-jetson-io 32.4.2-20200408182620 arm64 NVIDIA Jetson.IO debian package
ii nvidia-l4t-jetson-multimedia-api 32.4.2-20200408182620 arm64 NVIDIA Jetson Multimedia API is a collection of lower-level APIs that support flexible application development.
ii nvidia-l4t-kernel 4.9.140-tegra-32.4.2-20200408182620 arm64 NVIDIA Kernel Package
ii nvidia-l4t-kernel-dtbs 4.9.140-tegra-32.4.2-20200408182620 arm64 NVIDIA Kernel DTB Package
ii nvidia-l4t-kernel-headers 4.9.140-tegra-32.4.2-20200408182620 arm64 NVIDIA Linux Tegra Kernel Headers Package
ii nvidia-l4t-multimedia 32.4.2-20200408182620 arm64 NVIDIA Multimedia Package
ii nvidia-l4t-multimedia-utils 32.4.2-20200408182620 arm64 NVIDIA Multimedia Package
ii nvidia-l4t-oem-config 32.4.2-20200408182620 arm64 NVIDIA OEM-Config Package
ii nvidia-l4t-tools 32.4.2-20200408182620 arm64 NVIDIA Public Test Tools Package
ii nvidia-l4t-wayland 32.4.2-20200408182620 arm64 NVIDIA Wayland Package
ii nvidia-l4t-weston 32.4.2-20200408182620 arm64 NVIDIA Weston Package
ii nvidia-l4t-x11 32.4.2-20200408182620 arm64 NVIDIA X11 Package
ii nvidia-l4t-xusb-firmware 32.4.2-20200408182620 arm64 NVIDIA USB Firmware Package

This is very strange. Can you please dump output of below command on all boards and share here:

sudo /usr/sbin/nvpmodel -q

Last resort would be to move to latest JP 4.4.1.

Thanks,
Sharad

I was on a business trip. Sorry for the late reply,

The result ( both devices are the same)

NV Fan modeL quiet
NV Power mode: MODE_15W_6CORE
2

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

@PeterYu

We tried JP 4.4 DP image on multiple Jetson-Xavier NX boards, but still could not reproduce the issue. We need some additional details for further debugging.

clock tree dump:
cat /sys/kernel/debug/bpmp/debug/clk/clk_tree

APE Clock setting:
cat /sys/kernel/debug/bpmp/debug/clk/ape/min_rate
cat /sys/kernel/debug/bpmp/debug/clk/ape/max_rate
cat /sys/kernel/debug/bpmp/debug/clk/ape/rate
cat /sys/kernel/debug/bpmp/debug/clk/ape/possible_parents
cat /sys/kernel/debug/bpmp/debug/clk/ape/parent

APE Parent Clock setting:
cat /sys/kernel/debug/bpmp/debug/clk/<ape_parent>/min_rate
cat /sys/kernel/debug/bpmp/debug/clk/<ape_parent>/max_rate
cat /sys/kernel/debug/bpmp/debug/clk/<ape_parent>/rate
cat /sys/kernel/debug/bpmp/debug/clk/<ape_parent>/parent

where ape_parent is the output of “cat /sys/kernel/debug/bpmp/debug/clk/ape/parent”