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/TMS320F28020: Distribute software to an array of Piccolos

Part Number: TMS320F28020

Tool/software: Code Composer Studio

Hello,

I have to build a test bench with an array of piccolo processors which will all run the same program.The basic question is, how do I distribute the software to all of them as fast as possible? Debugging ist not necessary in the final application only during the first tests. But the final test bench should still be capabale to distribute new program versions within a minute fully automated.

The test bench will have several processors close to each other. Lets called it a cluster. My first idea was to chain the processors inside the cluster and use JTAG for programming. JTAG would be realised via USB cable and XDS debugger. Further, there will be more then one cluster. So this means with any additional cluster an additional USB cable would be necessary. The distance between PC and clusters does not prevent the use of USB cables, but it sounds bulky.

In my mind there should be a way to adress the clusters via ethernet. This would reduce at least the cable amount to the PC. So I found the XDS220 with ethernet capability, which should realise such functionallity. But additional 500 $ for each cluster to programm the piccolos is more then I can affort. Is a parallel programming with the help of those debuggers even possible? Is somone aware of a more cost effective way?

Additionally, the cluster chain may contain additional none TI chips that need software via JTAG too. Is it possible to add this software to the programming process started in CCS. In the end, maybe all the cluster chips will be programmed during one JTAG access interval.

I am still at the beginning of the construction and would be greatful for any experienced advice.

Kind regards,
Martin

  • Martin,

    In terms of a parallel programming soln we sell a 3rd party programmer on TI.com http://www.ti.com/tool/c2000-gang This allows up to 8 devices to be programmed in parallel either connected to a PC, or standalone(the flash image is stored on the programmer).

    If you wanted to do somthing with one emulator and many MCUs, that would be possible by hooking up the emulation ports in a daisy chain configuration(HW info here: http://dev.ti.com/tirex/#/?link=Development%20Tools%2FDebug%20Probes%2FXDS%2FDocuments%2FXDS%20Target%20Connection%20Guide; where each device would have its own JTAG ID.  You could then use the debug server that installs with CCS to program each device in the chain.  Debug server info is here in the CCS install path: C:\ti\ccsv8\ccs_base\scripting\docs 

    Finally, CCS could handled multiple TI devices in the way above; but if there are non-TI devices in the chain you may need something above CCS and whatever other IDE is needed for the other device to call its programmer.  

    Please let me know if you have more questions/comments and I'll be happy to look into them.

    Best,

    Matthew

  • Martin,
    Wanted to follow up to see if the above resolved your inquiry or if we need to keep this open. Will mark as TI thinks resolved, but feel free to reply back and it will re-open.

    Best,
    Matthew