UPGRADE YOUR BROWSER

We have detected your current browser version is not the latest one. Xilinx.com uses the latest web technologies to bring you the best online experience possible. Please upgrade to a Xilinx.com supported browser:Chrome, Firefox, Internet Explorer 11, Safari. Thank you!

cancel
Showing results for 
Search instead for 
Did you mean: 
Highlighted
Observer dblogicdesign
Observer
981 Views
Registered: ‎09-05-2018

[Labtools 27-3165] error, but FPGA is programmed?!

Jump to solution

I am using the VC707 Eval board.  When I power-up the board using the default FPGA configuration, the HW Manager connects and I can pull up the XADC dashboard.  All seems well. 

When I re-program the device with my own bitstream, I get the "[Labtools 27-3165] End of startup status:  LOW" error.  However, the DONE LED is lit, and I know my design is loaded since I added logic that flashes two of the user LEDs in a pattern.  This also means the clock is present.  There is only one clock (100 MHz - derived from on-board 200 MHz system clock) in my design and it also goes to the ILA I added to the design.  Changing JTAG clock didn't help.  Refreshing the device results in "[Labtools 27-2312] Device xc7vx485t_0 is no longer available." error.  Disconnecting and reconnecting the server, and then trying Auto Connect results in "[Labtools 27-2269] No devices detected on target localhost:3121/xilinx_tcf/Digilent/210203A3D1ADA" error.

Somehow programming my own device causes the device to disappear from the HW Manager.  I know the USB cable is good since it works with the default design.  SW11 DIP switch is set to JTAG mode.  Laptop connected to VC707 is running Win10 with Vivado 2018.2.  Any help is appreciated.

 

 

 

0 Kudos
1 Solution

Accepted Solutions
Observer dblogicdesign
Observer
866 Views
Registered: ‎09-05-2018

Re: [Labtools 27-3165] error, but FPGA is programmed?!

Jump to solution

Found the problem.  Simple mistake, in retrospect, but I will post what happened in case anyone else falls into this trap. 

I have a lot of I/O in my design that I assigned to either unconnected pins or the pins connected to the VC707's FMC connectors.  For my purposes, it is OK that the I/O are not connected, and I wanted to make sure they were not driving any of the other components on the board.  So I just assigned pins using the tables for the FMC connector in the VC707 User Guide (UG885). 

However, two of these signals are connected to switches that can disconnect the FPGA from the JTAG chain (see Fig 1.8, pp. 27, UG885 v1.7.1) if they are driven low.  These are named FMC1_HPC_PRSNT_M2C_B_LS and FMC2_HPC_PRSNT_M2C_B_LS (pins AM31 and AG32, respectively).  I reassigned my I/O away from these pins and everything worked.

6 Replies
Xilinx Employee
Xilinx Employee
965 Views
Registered: ‎06-06-2018

Re: [Labtools 27-3165] error, but FPGA is programmed?!

Jump to solution

Hi @dblogicdesign,

1. Ensure all the voltage rails all above respective Threshold levels and probe all the rails and ensure that there is no dip in any of the rails at start up and during configuration.

    Ensure all clock and data lines are free from Signal Integrity issues.

2. Please try reinstalling the Drivers by referring this https://www.xilinx.com/support/answers/59128.html.

3. Provide more CCLK clock cycles(10 or more) after DONE pin goes high, till EOS is asserted.

4. Try reducing the JTAG clock frequency to 3Mhz and try configuring the device.

5. Share status Register values.

6. If you have another VC707 board try detecting and configuring the device and also try with other JTAG Cable.

 

Regards,

Deepak D N

--------------------------------------------------------------------------------------------------------------------

Please Reply or Give Kudo or Mark it as an Accepted Solution.

--------------------------------------------------------------------------------------------------------------------

 

0 Kudos
Moderator
Moderator
964 Views
Registered: ‎01-16-2013

Re: [Labtools 27-3165] error, but FPGA is programmed?!

Jump to solution

Hi,

As per your information there is free running clock and your design works on single clock i.e. 100MHz.

The only problem I see here is the ILA clock, ILA clock should be 2x of data clock (i.e. 200MHz).

Use 200MHz system clock and get two clocks as output using MMCM/PLL i.e. 100MHZ (divide by 2) and 200MHz (i.e. divide by 1). Run your design using 100MHz clock and provide 200MHz (MMCM generated clock) to ILA clock input.

Let me know if this helps.

Thanks,
Yash

 

 

0 Kudos
Observer dblogicdesign
Observer
933 Views
Registered: ‎09-05-2018

Re: [Labtools 27-3165] error, but FPGA is programmed?!

Jump to solution

Thanks for the suggestion Yash.  I tried it, although I don't quite understand the reasoning.  All my signals going to the ILA are clocked by my 100 MHz clock.  This is the clock domain I chose for these signals in the wizard when adding my ILA.  I created a 200 MHz clock out of the MMCM I was using, and chose it as the domain for the debug signals, but I didn't see how this would help.  And it didn't.

0 Kudos
Observer dblogicdesign
Observer
930 Views
Registered: ‎09-05-2018

Re: [Labtools 27-3165] error, but FPGA is programmed?!

Jump to solution

Here is what happens when I re-program the device:

set_property PROBES.FILE {.....model_3.ltx} [get_hw_devices xc7vx485t_0]
set_property FULL_PROBES.FILE {.....model_3.ltx} [get_hw_devices xc7vx485t_0]
set_property PROGRAM.FILE {.....model_3.bit} [get_hw_devices xc7vx485t_0]
program_hw_devices -disable_eos_check [get_hw_devices xc7vx485t_0]
program_hw_devices: Time (s): cpu = 00:00:55 ; elapsed = 00:00:55 . Memory (MB): peak = 3160.941 ; gain = 0.000
refresh_hw_device [lindex [get_hw_devices xc7vx485t_0] 0]
ERROR: [Labtools 27-2312] Device xc7vx485t_0 is no longer available.
Check cable connectivity and that the target board is powered up then
use the disconnect_hw_server and connect_hw_server to re-initialize the hardware target.
Use open_hw_target to re-register the hardware device.
ERROR: [Labtoolstcl 44-513] HW Target shutdown. Closing target: localhost:3121/xilinx_tcf/Digilent/210203A3D1ADA

As I stated before, disconnecting from the server and re-connecting doesn't work.  I still can't see the device.  The above was an attempt with JTAG clock at 3 MHz.  I've tried two different cables.  Both can access the board fine BEFORE I download my bit file.  Again, the device is getting programmed, as I see the two user LEDs flashing per my design, yet the device disappears from the JTAG chain.

0 Kudos
Observer dblogicdesign
Observer
892 Views
Registered: ‎09-05-2018

Re: [Labtools 27-3165] error, but FPGA is programmed?!

Jump to solution

I am still trying to figure out what's going on here.  I have explored various ideas and tried many things, but I still am experiencing this issue.  I can power up the VC707, see the target FPGA (including the XADC) in the HW manager, but as soon as I program it I get the errors I posted previously and the FPGA "disappears" from the JTAG chain, and nothing I do seems to be able to bring it back. 

I suspect there must be some setting when compiling a custom design for the VC707 that, if set incorrectly, somehow "kills" the JTAG connection after the device is programmed.

Help!

0 Kudos
Observer dblogicdesign
Observer
867 Views
Registered: ‎09-05-2018

Re: [Labtools 27-3165] error, but FPGA is programmed?!

Jump to solution

Found the problem.  Simple mistake, in retrospect, but I will post what happened in case anyone else falls into this trap. 

I have a lot of I/O in my design that I assigned to either unconnected pins or the pins connected to the VC707's FMC connectors.  For my purposes, it is OK that the I/O are not connected, and I wanted to make sure they were not driving any of the other components on the board.  So I just assigned pins using the tables for the FMC connector in the VC707 User Guide (UG885). 

However, two of these signals are connected to switches that can disconnect the FPGA from the JTAG chain (see Fig 1.8, pp. 27, UG885 v1.7.1) if they are driven low.  These are named FMC1_HPC_PRSNT_M2C_B_LS and FMC2_HPC_PRSNT_M2C_B_LS (pins AM31 and AG32, respectively).  I reassigned my I/O away from these pins and everything worked.