• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 97 subscribers
  • Views 461 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.

RTDX vs. Trace

Jyri Öörni
Prodigy 30 points

Hello,

We are in a process of choosing the platform for an embedded control system. C6000 floating point DSPs seem like a suitable option, but unfortunately I don't have any experience on TI DSPs or development tools. We are also trying to get an idea of the debugging tools, and one interesting area are the RTDX and the Trace features. I know RTDX is a software way of exchanging data between target and host and that Trace is a hardware solution, but how do they differ in functionalities?

As I told I don't have experience on either one, but I do have experience on the Chipscope feature on Xilinx FPGA environment, where you can add a debugger core in your application and capture a buffer with number of signals from certain number of cycles. You can configure trigger conditions that trigger the capture and set the debugger wait for a triggering. When the buffer is captured, transfer of the data up to the host PC is initiated. On the host application you can view the data in graphical form as waveforms a bit like on an oscilloscope.

Does either one implement same kind of functionalities?

 

Thanks

  • 0
    TI__Mastermind 18130 points

    Hi Flotari,

    RTDX is an acronym for Real-Time Data eXchange. This is just a fancy term for passing data between the target and host without halting the target's processor. All RTDX communication travels between host/target via your emulator connection.

    Trace is a more advanced method of debugging in that it actually maintains a buffer containing previously executed code. You can find a much more in-depth discussion of this in our eXpressDSP article. Please also note that because the Trace emulator requires additional hardware on the device that this is limited to very few DSP parts, none of which are floating-point.

  • 0 in reply to TimHarron
    Prodigy 30 points

    Hi Tim,

    Thanks for the quick answer!

    Allright, so Trace is not an option for the floating point C6000. But RTDX could be used for the kind of debugging that Chipscope provides in the FPGA environment? How about the resource usage of RTDX? I know it always depends but in general is it light or heavy?

     

    Thanks

  • 0 in reply to Jyri Öörni
    TI__Mastermind 18130 points

    I have not worked with ChipScope before so I went and found this description from the Xilinx website:

    ChipScope™ Pro tool inserts logic analyzer, bus analyzer, and virtual I/O low-profile software cores directly into your design, allowing you to view any internal signal or node, including embedded hard or soft processors.
    This is going to be beyond the scope of what RTDX handles. RTDX is basically capable of simple data buffer transfers between the host and target whereas it looks like ChipScope is a full-featured bus analyzer inside the FPGA. Our DSPs would need a specific bus analyzer module built into the silicon (which would cost more and increase die size).

    With regards to resource usage this should be invisible to the DSP, so resource usage is minimal. The DSP will make a function call and effectively be done with it.

  • 0 in reply to TimHarron
    Prodigy 30 points

    Allright, thanks a lot.

    What we're mostly interested in IS the data transfer between target and host. Unclear at the moment is what initiates the transfers and that can the conditions for this be set remotely somehow. Or does all this have to be hand coded in the target application?