Is my XC-2 or XTAG Dead?

[jb007]

Hi all, as the title suggests I'm having problems programming my XC-2 using the XTAG FTDI device. It was working until recently, but I get the following error from the XDE when run as an "XCore Application":

Code: Select all

xrun: Problem in connection to device 

and as a "Write to Flash Memory"

Code: Select all

Error: F03034 Failed to fully run flash inquisitor.
xrun: Problem in connection to device

At first I thought it was related to my OS and FTDI drivers, but alas it's not; I've tried it under Windows 7 64-bit, Windows 7 running XP SP3 under Windows Virtual PC, and Kubuntu 10.10 using the patched xftd2xx.dll that XMOS support sent me.

If I do a device listing I get:

Code: Select all

jb@i7:~/XMOS/DesktopTools/10.4.2$ xrun -l

Available XMOS Devices
----------------------

  ID    Name                    Adapter ID      Devices
  --    ----                    ----------      -------
  0     XMOS XTAG               XMS2OWT0        None

jb@i7:~/XMOS/DesktopTools/10.4.2$ 

Notice that it can see the XTAG, but not the XC-2 (Devices = None). Next I fired up my DSO and looked at the JTAG signals on the XTAG header J2. I can see data being clocked into the XC-2 on Pin 5 TDSRC (TDI) and Pin 9 TCK, but nothing is coming back on Pin 13 TOSNK (TDO). All voltage levels are fine 5, 3.3 and 1.

Does anyone have any constructive clues as to what to look for next? I'm pulling out what is left of my hair! :?

Jim

[leon_heller]

I've got a spare XTAG I could loan you (I've got two XC-2s).

[XMatt]

Hi all, as the title suggests I'm having problems programming my XC-2 using the XTAG FTDI device. It was working until recently, but I get the following error from the XDE when run as an "XCore Application":

Code: Select all

xrun: Problem in connection to device 

and as a "Write to Flash Memory"

Code: Select all

Error: F03034 Failed to fully run flash inquisitor.
xrun: Problem in connection to device

At first I thought it was related to my OS and FTDI drivers, but alas it's not; I've tried it under Windows 7 64-bit, Windows 7 running XP SP3 under Windows Virtual PC, and Kubuntu 10.10 using the patched xftd2xx.dll that XMOS support sent me.

If I do a device listing I get:

Code: Select all

jb@i7:~/XMOS/DesktopTools/10.4.2$ xrun -l

Available XMOS Devices
----------------------

  ID    Name                    Adapter ID      Devices
  --    ----                    ----------      -------
  0     XMOS XTAG               XMS2OWT0        None

jb@i7:~/XMOS/DesktopTools/10.4.2$ 

Notice that it can see the XTAG, but not the XC-2 (Devices = None). Next I fired up my DSO and looked at the JTAG signals on the XTAG header J2. I can see data being clocked into the XC-2 on Pin 5 TDSRC (TDI) and Pin 9 TCK, but nothing is coming back on Pin 13 TOSNK (TDO). All voltage levels are fine 5, 3.3 and 1.

Does anyone have any constructive clues as to what to look for next? I'm pulling out what is left of my hair! :?

Jim

It looks like the tools cannot talk to the chip correctly, this suggests that there is a problem with your XC-2 board, I suggest you contact the official XMOS support team to debug this problem further and replace the board if required. You can contact them at support@xmos.com.

[jb007]

I've got a spare XTAG I could loan you (I've got two XC-2s).

Thanks for the offer, but I'm located in Melbourne Australia... thanks anyway. :)

[segher]

You say you see no activity on TDO (out as seen from the chip). Did you test this
on the connector? You can try to see if there is some test point closer to the chip;
it might just be a bad contact on the connector. If not, it sounds like you fried the
JTAG on the chip :-(

[jb007]

You say you see no activity on TDO (out as seen from the chip). Did you test this
on the connector? You can try to see if there is some test point closer to the chip;
it might just be a bad contact on the connector. If not, it sounds like you fried the
JTAG on the chip :-(

I measured TDO at the header pin 13, it measures a constant 3.3V. It goes straight to the XS1-G4, no test points.

Doing further test's with DSO, (wish I had a logic analyser), looks like the clock from the XTAG is clocking (excuse pun) but the TDI pin just goes high for the entire period that the clock clocks ie all 1's during clocking. This might explain no output from TDO?

[segher]

I measured TDO at the header pin 13, it measures a constant 3.3V. It goes straight to the XS1-G4, no test points.

Doing further test's with DSO, (wish I had a logic analyser), looks like the clock from the XTAG is clocking (excuse pun) but the TDI pin just goes high for the entire period that the clock clocks ie all 1's during clocking. This might explain no output from TDO?

Do you see anything on TMS? The standard way to identify the devices on the JTAG
chain is to clock in only ones to the DR reg, right after test reset, so seeing only ones on TDI
doesn't conclusively mean it is bad. You should see both zeroes and ones on TMS though.

You could try hooking up some other JTAG master, if you have one, perhaps.

[jb007]

I measured TDO at the header pin 13, it measures a constant 3.3V. It goes straight to the XS1-G4, no test points.

Doing further test's with DSO, (wish I had a logic analyser), looks like the clock from the XTAG is clocking (excuse pun) but the TDI pin just goes high for the entire period that the clock clocks ie all 1's during clocking. This might explain no output from TDO?

Do you see anything on TMS? The standard way to identify the devices on the JTAG
chain is to clock in only ones to the DR reg, right after test reset, so seeing only ones on TDI
doesn't conclusively mean it is bad. You should see both zeroes and ones on TMS though.

You could try hooking up some other JTAG master, if you have one, perhaps.

I think tms does toggle, but I'll check tomorrow as it's late here in oz. I have a Xilinx (Digilent) JTAG, could that be made to work? Thanks for your help.

[leon_heller]

If it's a USB JTAG and uses the same FTDI chip it might work. You will need an adapter if the JTAG signals are on different pins.

[jb007]

I measured TDO at the header pin 13, it measures a constant 3.3V. It goes straight to the XS1-G4, no test points.

Doing further test's with DSO, (wish I had a logic analyser), looks like the clock from the XTAG is clocking (excuse pun) but the TDI pin just goes high for the entire period that the clock clocks ie all 1's during clocking. This might explain no output from TDO?

Do you see anything on TMS? The standard way to identify the devices on the JTAG
chain is to clock in only ones to the DR reg, right after test reset, so seeing only ones on TDI
doesn't conclusively mean it is bad. You should see both zeroes and ones on TMS though.

You could try hooking up some other JTAG master, if you have one, perhaps.

If it's a USB JTAG and uses the same FTDI chip it might work. You will need an adapter if the JTAG signals are on different pins.

Thanks Seger & Leon, yes I am getting toggling on TMS. I have a Digilent Nexys2 FPGA development board, and I made an adapter cable to go between the XTAG & the JTAG port on the Nexys2.

When I run:

Code: Select all

C:\Program Files (x86)\XMOS\DesktopTools\10.4.2>xrun -l

Available XMOS Devices
----------------------

  ID    Name                    Adapter ID      Devices
  --    ----                    ----------      -------
  0     XMOS XTAG               XMS2OWT0        ??None


C:\Program Files (x86)\XMOS\DesktopTools\10.4.2>

Notice the two question marks before the Device; I assume that it puts them there because the device returned is different to expected. I also see a lot of activity from TDO.

My assumption is that something is wrong with the XC-2, not with the XTAG or desktop drivers and or software.

Bummer! I will continue to keep looking, focusing on the XC-2... maybe I should buy another whilst the Aussie dollar is strong.