• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 101 subscribers
  • Views 2007 views
  • Users 0 members are here
Support feedback
Options
Options
Similar topics

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.

Received exception from ROV server

Magne Pettersen
Prodigy 130 points

Hello,

I have just upgraded a fully functioning project from using CCSv4 to CCSv6, and am now experiencing problems with heap.

Target is c5509A, using DSP/BIOS 5.42.0.07 and CGT 4.4.1

The project builds and the debugger reaches main().

In ROV->MEM, I see the error: "Received exception from ROV server: Target memory read failed at address: 0x0, length: 4"

Now, heap has been defined in the .tcf-file (along with stack and sysstack). New from CCSv4 (DSP/BIOS 5.41.10.36) is that these sections must also be defined in the Linker options. Defining --heap_size in the Linker options, creates a memory segment '.sysmem' (whichever size I set it to be). I thought the heap definition in the .tcf-file would override the Linker options?

And I am not allowed to not create a heap in .tcf (using the graphical config tool).

Any help is greatly appreciated,

Magne

  • 0
    TI__Mastermind 21645 points
    Hi Magne,

    Are you only seeing issues with ROV view for MEM module or is the application also breaking ?

    Best,
    Ashish
  • 0 in reply to Ashish Kapania
    Prodigy 130 points

    Hello Ashish,

    I can single step through the main() function ok, but as soon as the scheduler takes over the program is lost.

    Magne

  • 0 in reply to Magne Pettersen
    TI__Mastermind 21645 points

    Hi Magne,

    Magne Pettersen said:

    I can single step through the main() function ok, but as soon as the scheduler takes over the program is lost.

    Ok, so the application is also not behaving as expected.  Why I asked if the application was running properly was to determine if it is actually an issue with ROV or the app was somehow corrupted that caused ROV's MEM view to stop working. It seems the app has a problem which is causing ROV to not work.

    DSP/BIOS v5 is a legacy product so its going to be hard to debug this but we can try. You mentioned the program is lost after the scheduler takes over. What is the CPU running/doing if you halt the core ? And also, do you get any error msgs on the console ?

    Best,

    Ashish

  • 0 in reply to Ashish Kapania
    Prodigy 130 points
    Ashish, thanks for the reply,

    Once the scheduler takes over the program seems to run on random code. If I halt the core, it just stops in a random location with random code. Sometimes the connection to the debugger is lost and an error dialog pops up.
    No errors on the console though.

    I did receive a call from TI support (due to this thread). They said that CCS v6 only partially supports the C5000-series, and that we should try using CCS v5.5 instead (which has better support for C5000). [the reason for the upgrade was to be able to develop on both C5509 and newer C55x processors]
    Tried v5.5 yesterday, and it seems to work fine; debugging, ROV etc. (a minor issue from the migration process remains though).

    As I understand, further development on C5x (and DSP/BIOS v5) is not such a hot topic for CCS v6. Since CCS v5.5 seems to provide the functionality we need, that will most likely be 'the solution' for us.

    Best Regards,
    Magne