• State TI Thinks Resolved
  • Locked Locked
  • Replies 1 reply
  • Answers 1 answer
  • Subscribers 99 subscribers
  • Views 273 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

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.

TDA4AEN-Q1: Why the runtime of the post-processing customize kernel is abnormal relative to the tda4VM ?

linxia yang
Prodigy 40 points
Part Number: TDA4AEN-Q1
Other Parts Discussed in Thread: TDA4VM

Tool/software:

Hi,

I have a customize kernel that runs the post-processing of the similar YOLOX model, On TDA4VM, the A72 cores and C66 cores run at 0.906ms and 9.122ms, respectively. Everything works fine!!

BUT.

When the same code is migrated on TDA4AEN, On TDA4AEN, the A53 cores and C7X cores run at 194ms and 545ms, respectively. It's super slow!

Excuse me, what is the possible problem ?

POSTPROC RUN ON  A72 CORE OF TDA4VM

POSTPROC RUN ON  C66 CORE OF TDA4VM

POSTPROC RUN ON  A53 CORE OF TDA4AEN

POSTPROC RUN ON  C7X CORE OF TDA4AEN

  • 0
    TI__Guru* 84926 points

    Hi

    The timing mentioned here is the execution time of the process callback of your custom kernel. 

    The contents inside the process callback is your custom implementation. I would suggest you to please review the same and analysis where the delay in execution time occurs inside this process callback.

    Regards,

    Nikhil