• Locked Locked
  • Replies 5 replies
  • Subscribers 119 subscribers
  • Views 3720 views
  • Users 0 members are here
Options
Related
How was your experience today?
This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Unable to count event in Cortex R4 using PMC

sandesh mukartihal
Note: This was originally posted on 22nd July 2011 at http://forums.arm.com

Hi,

I am trying to set the event counter register to count particular event using ARM Cortex R4's performance monitor register.

I followed the steps as in this link..
http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.faqs/ka4237.html

But it is not counting the events which I set. Only the cycle count is working. I am using RVDS 4.1 prof.

Does RVDS support event counting or is there any configuration error that needs to be corrected?
  • Note: This was originally posted on 25th July 2011 at http://forums.arm.com

    Hi,

    I am using the ISSM. Yes it just models the cycle count but not the PMU.

    Is there any other tools/simulators available that model PMU for any cortex R or any cortex A series processors?

    Thank You.


    What target are you connecting to? 

    If you are using the models what ship with RVDS, these do not model the PMU counters (except the CCNT - I think).
  • Note: This was originally posted on 28th July 2011 at http://forums.arm.com

    Hi,

    If I use a beagle board as target, will I be able to use the PMC's then?, will the RVDS update the PMC register if I connect to beagle board and program it?

    Thank you for your valuable time.


    What target are you connecting to? 

    If you are using the models what ship with RVDS, these do not model the PMU counters (except the CCNT - I think).
  • Note: This was originally posted on 28th July 2011 at http://forums.arm.com

    Thank you very much.

    I actually want to measure the branch prediction related events accurately for my code and I don't want any background application/process to be running when I run my code. So thereby I'll be getting accurate branch related counter info. Is it possible to use beagle board in this configuration and debug target using RVDS?

    I have a panda board with dual cortex A9 in it, but it has to be loaded with boot loader and linux OS. Doing so I cant measure PMC's accurately for my program if I run it on OS. Am I right regarding this or is there a way to go about it? Considering this, and the fact that ARM workbench supports beagle board target I am thinking of switching to beagle board.

    Any suggestions will be of great help to me. Thank you for your time.

    If you use real hardware (a board), then yes you will have the PMU present.  Remember that by default the PMU is only accessible in privileged modes.  User mode access must be manually set if you required.
  • Arm Employee Badge
    Note: This was originally posted on 25th July 2011 at http://forums.arm.com

    What target are you connecting to? 

    If you are using the models what ship with RVDS, these do not model the PMU counters (except the CCNT - I think).
  • Arm Employee Badge
    Note: This was originally posted on 28th July 2011 at http://forums.arm.com

    If you use real hardware (a board), then yes you will have the PMU present.  Remember that by default the PMU is only accessible in privileged modes.  User mode access must be manually set if you required.