If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Flash prebuilt 8.4 EdgeAI SDK for SK-TDA4VM onto SD card Add a printf statement in the main function. Specifically: Run the following commands to build and copy over file to SD card
1999 cd vision_apps/ 2000 make sdk 2001 sudo cp out/J7/A72/LINUX/release/vx_app_arm_ipc.out /media/anzhigang/rootfs/opt/ 2002 history
Then,Two more questions，Need help answering： 1、I added the print in your patch file and can see the output log normally？ 2、log output of the debug version is the same as that of the realese version？
You should see the exact same output as I was seeing. If the software is the same and the hardware is also the same, then I do not have a good idea on why there is a difference between our setup.
The debug version should have more print statements, but the printf the expert added in the patch should work without building for the debug version.
vx_app_arm_ipc.out- we attached vx_app_arm_ipc.out, but can you try running this? If you see "Testing printf" at the start of the logs, then there is something that is wrong with how printf is added to the code, or something wrong with the build. If there are no "Testing printf", we would suspect something wrong with the underlying image flashed onto the SD card.
I described in my post from July 24th the steps I have ran to compile and run printf test. I have also copied the relevant logs that shows that printf is functioning as expected.
If you are following the steps exactly as described, then there should be no difference between our logs. However, looking through the steps that have been shared with me, itdoeslook like the steps are followed but resulting logs are different.
Because there are some differences in resulting behavior, I would like to make our setup as similar as possible by sharing my vx_app_arm_ipc.out that I have verified the extra printf is working.Please run the .out file in my previous post, and share logsso that we can narrow down whether the issue is in the underlying EdgeAI SDK image flashed onto the SD card, or an issue with the build environment/procedure.
