• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 54 subscribers
  • Views 899 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.

CCS/PROCESSOR-SDK-OMAPL138: Finding Clock cycle

AravindKrishnan M
Prodigy 110 points

Part Number: PROCESSOR-SDK-OMAPL138
Other Parts Discussed in Thread: OMAPL138

Tool/software: Code Composer Studio

I just want to check the number of clock cycles required for a part of my c code actually the code is in micro c os idle task. anyway the code is  running in the arm core of omap l138. I tried to use the clock in CCS but it shows context id instead of clock cycles.It shows clock cycle while i debug in the DSP core . I need to check in the arm core. How can I do that? 

  • 0
    TI__Guru**** 475221 points
    Hello,
    Yes I see that also. I suspect that ContextID is actually measuring clock cycles and that the mapping of the names under the hood is incorrect (we have had this happen in the past before). If that is the case, then that is a bug. I do need to first confirm that this is the case. I will reply back with my findings.

    Thanks
    ki
  • 0 in reply to Ki
    Prodigy 110 points
    Hello,
    Thank you for your reply.I too now suspect the same. What is the difference between enabling the clock and finding clock cycle by using count event in the break point section? I got different values.
  • 0 in reply to AravindKrishnan M
    TI__Guru**** 475221 points
    I tried this out on my OMAPL138 target and I get the same values. Both the profile clock and the event counter reads the cycles from the same register, hence it should be the same value. Note that when both prof clock and event counter is enabled, the prof clock will incrementally increase while the event counter will always reset to '0' before counting to the next halt point. This may explain the difference you see.

    Based on my analysis, it appears that contextID is really cycles. I will file a bug for the naming issue

    Thanks
    ki
  • 0 in reply to Ki
    Prodigy 110 points
    Hello
    Thank you for your reply.