Intel® Makers
Intel® Edison, Intel® Joule™, Intel® Curie™, Intel® Galileo
Welcome - This is a Peer-to-Peer Forum only. Intel has discontinued these products but you may find support from other customers on this Forum
9873 Discussions

Getting Code into Flash Area


I am using ISSM_2016.0.027 on Windows 7 64 bit

With the Eclipse IDE when I select Debug it is not updating the flash. It seems to have done it once because in the disassembly I can see an old version of the code loaded.

Nothing seems to be able to prompt Eclipse to reload new code.

So I thought I would try the command lin method way described in the eclipse doco "Intel® System Studio 2016 for Microcontrollers User and Reference Guide"

Where is says

  1. Open a terminal to run JTAG-OpenOCD: source
    1. ISSM _ROOT/
    2. cd $ISSM_DEBUGGER_ROOT/openocd
    3. bin/openocd -f scripts/board/quark_d2000_onboard.cfg
  2. Open a second terminal to connect GDB to port 3333 source
    1. ISSM_ROOT/
    2. cd $ISSM_DEBUGGER_ROOT/gdb
    3. ./gdb-ia target remote localhost:3333
    4. monitor gdb_breakpoint_override hard
    5. set remotetimeout 30
    6. monitor clk32M
    7. file my/elf/file/here.elf
    8. load
    9. monitor reset halt
    10. flushregs
    11. tb main c

Which is quite involved to say the least. But once I send it the "target remote localhost:3333" I get "localhost:3333: The system tried to join a drive to a directory on a joined drive."

If I plough on I eventually get a 'can't do that when target it exec ' when I try to load.

To be honest I am not so interested in the debugger - just some way of flashing the parts. It there any reliable method for programming available for windows?

Also - I do not have a crystal on the micro. Is the Jtag programming going to mind not having an external Crystal? The application uses the internal oscillator initially. The program ROM (for the OTP ROM) under Eclipse program reliably.

0 Kudos
4 Replies

Hi DonAlan,

I am sorry to hear you have some problems while programming the board with Windows. I would like to ask you some questions to identify the root of the issue and solve it.

1. Are you using the Quark D2000 microcontroller development kit?

2. Did you update the ROM first?

3. Could you add some screenshots of the error messages you get on Eclipse?

4. I'm not sure if you have already seem the /community/tech/microcontrollers/blog/2016/04/06/limitations-and-known-issues-in-intel-system-studio-2016-for-microcontroller? Limitations and Know Issues in the Intel System Studio 2016; as you can see in there, if you use the UART-B or PWM0 signals, this may produce some problems while uploading code.

5. If you are using the Quark D2000 microcontroller development kit, in the schematic you can see that the FT232H uses its own crystal.

Please let us know as much you can, so we could replicate your environment if possible.

Kind regards,



I am running the processor in our own hardware. The Jtag interface I use is a copy of what is on the dev kit. If I use the ISSM Update Rom on the Eclipse platform that works fine every time.

I have built my own code for the OTP rom area as the delivered code tries to switch to crystal osc.

That seems to work. The code executes to 0x180?054 in the flash where I have an endless loop. Trouble is if I change the code and go back to debugger it does not reflash the part. No errors. It just debugs with old code loaded. I took the other projects out of the work space and reloaded eclipse etc.

As regard 4 i am aware of the potential to brick the device but I think if that had hapenned the debugger would not be working at all. Debugging and prog OTP rom are fine. I just cant find what stimultes Eclipse to try to program Flash.




What differences has your code from the rom_startup.c for the D2000? Could you share the lines where you did the changes?

Also, could you share some schematics of the hardware you are using?.





Thanks. I spent a couple of days writing code and went back to this issue. Today it is working OK. Maybe in my impatience I did not let it hand long enough at 93% complete as it programs the flash. So all good currently. Sorry to trouble you.