Forum Discussion

Altera_Forum's avatar
Altera_Forum
Icon for Honored Contributor rankHonored Contributor
20 years ago

Problem Downloading in Nios IDE

I am trying to program is simple "Hello World" program into a Nios II running in a Cyclone (EP1C6T144) which is on a custom board. I'm using Quartus 5.0 and the USB byte-blaster. When I select "Run As - Nios II hardware" I get the following message.

===

There are no Nios II CPUs with debug modules available which match the values

specified. Please check that your PLD is correctly configured, downloading a

new SOF file if necessary.

===

I'm able to download the .sof file from the Quatus II programmer just fine so I'm pretty sure that the JTAG is working well. I made sure that the IDE project points to the right .ptf file and the Nios II in SOPC has a level 1 debug module. What values is the error message referring to? I tried taking out the system ID from SOPC Builder.

I read on another posting that there might be an issue with the TCK line and that a simple RC with 100pF and 32ohm might help. I tried that without success. Can you think of what might be stopping the .elf file from being downloaded?

Thanks,

Danny

6 Replies

  • Altera_Forum's avatar
    Altera_Forum
    Icon for Honored Contributor rankHonored Contributor
    20 years ago

    Hi Danny,

    My two cents would be to:

    1. Try a JTAG if possible and see if you get the same message

    2. Try rebooting your machine, sounds hokey, but I have seen this work

    3. Try a different CPU, Like The Lower Performance/ Also play wiht debug levels

    4. Make sure memory is targeted correctly, check system properties.

    -Baycool
  • Altera_Forum's avatar
    Altera_Forum
    Icon for Honored Contributor rankHonored Contributor
    20 years ago

    It looks, to me, like you've not programmed your FPGA or you have no debugger built into your processor, though you clearly state that you have. What does a "jtagconfig -n" command return from within the SDK shell? You should see something like this: 

    $ jtagconfig -n
1) USB-Blaster 
  020840DD   EP1C20
    Node 11104600
    Node 0C006E00

    The two "nodes" listed, above, are the debugger and the jtag_uart, in my case.

    Also, make sure you're not using an OCP version of the processor, as that requires that you keep the programmer open.

    Cheers,

    - slacker
  • Altera_Forum's avatar
    Altera_Forum
    Icon for Honored Contributor rankHonored Contributor
    20 years ago

    I've tried everything that Baycool suggested and still nothing. I've put together a second board and it is acting the same way. When I did the jtagconfig -n command in the SDK shell I get the first two lines correctly followed by 30 nodes! I only have one Cyclone FPGA and as far as I know I only have one Nios II. What could cause that?

    I've been fighting this Nios problem for a while now. When I first tried to program the Cyclone through the JTAG port, the Quartus II programmer would not program the Cyclone and return the message that "Config_done does not go high". Only after I programed a .pof file into the EPCS flash (thereby loading the configuration using active serial) was I able to program the cyclone through the JTAG. Could these two seemingly unrelated problems have a common link?

    Rgds,

    Danny
  • Altera_Forum's avatar
    Altera_Forum
    Icon for Honored Contributor rankHonored Contributor
    20 years ago

    You've got JTAG hardware issues. Take a look at one of the Nios II Dev. board's schematic.... It should steer you in the right direction.

    Cheers,

    - slacker
  • Altera_Forum's avatar
    Altera_Forum
    Icon for Honored Contributor rankHonored Contributor
    20 years ago

    I layout the JTAG as:

    TCK pulled low with 10K ohm

    TMS,TDI pulled high with 10K ohm

    TDO floated ( I saw someone pllued high)

    other signals for downloads:

    FPGA_CONF_DONE,FPGA_STATUSn pulled high with 10K ohm

    I tested TDO pllued high, It worked, too. Which one is the best choice?
  • Altera_Forum's avatar
    Altera_Forum
    Icon for Honored Contributor rankHonored Contributor
    20 years ago

    Thank everyone for all of the suggestions. It turns out that I did have a JTAG problem as some of you pointed out. It wasn't a schematic issue but rather a problem with my layout. The solution ended up being a couple of decoupling caps right next to the JTAG header power pins. I used 1uF in parallel with 1000pF. Apparently I was getting some noise in the JTAG signals. I haven't had any problems since.

    Danny