Forum Discussion

dpcsu's avatar
dpcsu
Icon for New Contributor rankNew Contributor
1 month ago
Solved

jtagserver.exe causing BSOD together with ftdi driver

Dear developers in the same shoe like me,

I am facing a strange issue, and would like to ask others whether they have similar problem:

Quartus Prime Pro 25.3.0 was installed first and in use with USB Blaster III (evalkit)

After some time (probably windows pulled some newer driver components) each time when I plugged in the evalboard it was crashing the system.

With WinDBG analyzing the minidump it turned out the responsible was jtagserver.exe

calling some unknown function in ftdibus.sys

"PROCESS_NAME: jtagserver.exe

SYMBOL_NAME: ftdibus+f4d4

MODULE_NAME: ftdibus

IMAGE_NAME: ftdibus.sys

STACK_COMMAND: .cxr; .ecxr ; kb

BUCKET_ID_FUNC_OFFSET: f4d4

FAILURE_BUCKET_ID: 0x139_a_GUARD_ICALL_CHECK_FAILURE_ftdibus!unknown_function

OSPLATFORM_TYPE: x64

OSNAME: Windows 10 --> this is windows 11

FAILURE_ID_HASH: {3a7c8e82-5c61-9acf-ceb5-61656fa0b04c}"

And here comes the strange part:

Reinstalling the driver with dp_inst.exe from Quartus directory helped until next restart or at least for 1-2 hours. Then power cycling evalboard caused a crash again.

When I fresh reinstalled driver and plugged in again the evalboard again everything was fine.

When I updated to the latest quartus, the driver package in it was causing crash again.

And I mean here reinstalling driver from the 25.3.1 quartus version did not change a thing.

 

Anybody facing similar issue: I recommend installing Windows SDK and use WinDBG there for crashdump analysis to confirm this.

 

I am looking forward for any feedback or stable solution idea.

 

Kind regards,

Peter

general
Installation & Licensing
  • RichardT_altera's avatar
    RichardT_altera
    1 month ago

    Does the BSOD still occur if you never power‑cycle the eval board?

    I’m asking because the behavior you’re seeing is consistent with a mixed‑FTDI driver environment, which is a plausible root cause.
    Even though the USB‑Blaster III uses an Intel‑signed INF with a custom VID/PID, both Intel’s package and the standard FTDI CDM packages typically load the same kernel‑level bus driver: ftdibus.sys. This SYS file is shared across many FTDI‑based devices.

    If Windows has multiple FTDI packages staged (e.g., one from Intel’s USB‑Blaster III and others for FT2232H/FT4232H products), the system can:

    -Select a different FTDI package on re‑enumeration (after a reboot or cable power cycle).
    -End up loading a different ftdibus.sys build (older or incompatible) for the same hardware instance.

    This aligns with your observation that a clean reinstall temporarily resolves the issue, but the crash returns later.
    Could you check whether multiple versions of ftdibus.sys exist on the machine?

    Regards,
    Richard Tan

6 Replies

  • dpcsu's avatar
    dpcsu
    Icon for New Contributor rankNew Contributor
    30 days ago

    hi Richard

    so I removed all the usb blaster and all the ftdi drivers and reinstalled just the latest ones...
    so far it is ok. I will tell you tomorrow after 1 day of usage if it is still fine.

    Thanks for the hints!

    Kind regards,
    Peter

    • RichardT_altera's avatar
      RichardT_altera
      Icon for Super Contributor rankSuper Contributor
      26 days ago

      Hi Peter, 

      Hope you're doing well.
      Just checking in—has the issue remained resolved over the past days? 

      Please let me know if everything is still working fine or if you’ve noticed anything unusual.

      Regards,
      Richard Tan

      • dpcsu's avatar
        dpcsu
        Icon for New Contributor rankNew Contributor
        25 days ago

        Hi Richard,

        yes it works now.

  • RichardT_altera's avatar
    RichardT_altera
    Icon for Super Contributor rankSuper Contributor
    1 month ago

    I am currently suspecting that the issue may be caused by either driver incompatibility or a USB‑Blaster III hardware‑related problem.
    Could you help verify the following?
    1. Driver installation
    Have you installed the USB‑Blaster III drivers exactly as described in the user guide below, ensuring that both components were installed?

    USB‑Blaster III driver (requires UART FTDI driver)
    UART FTDI driver

    User Guide:
    https://docs.altera.com/r/docs/864466/current/usb-blaster-iii-fpga-development-cable-user-guide/connecting-the-usb-blaster-iii-fpga-development-cable

    2. Try installing only the Programmer
    Could you uninstall the entire Quartus Pro software suite and instead install only the:
    “Quartus® Prime Pro Edition Programmer and Tools”
    download link below (at "Individual Files" tab): https://www.altera.com/downloads/fpga-development-tools/quartus-prime-pro-edition-design-software-version-25-3-1-windows

    3. Test with only the USB‑Blaster III connected. Please disconnect all other USB devices and hardware.
    Then connect only the USB‑Blaster III to rule out interference from other peripherals.

    4. Test a different USB‑Blaster III unit
    Does the issue still occur if you try using another USB‑Blaster III cable?

    5. Test on another Windows 11 machine
    Does the same problem occur if you connect the USB‑Blaster III to a different Windows 11 PC?

    I may need some additional time to reproduce this issue on my side.

    Regards,
    Richard Tan

    • dpcsu's avatar
      dpcsu
      Icon for New Contributor rankNew Contributor
      1 month ago

      Hi Richard,

      thank you, for picking up this topic!

      1. yes, both were installed
      2. will try
      3. nothing else was connected before, just the Agilex3C evalboard, as soon as USB connection was made and JTAG server tried to communicate with it the system crashed
      4. It happens still, I tried with two evalboard
      5. It did not happen on a machine where Quartus was not installed (I know this is a too easy statement since jtagserver.exe was causing it according to minidump), I will try it on a different computer

        So summary: 2,5: will be tried, 1,3,4: did not help further.

       

      We have one more theory:

      We are using the machine with products where there is FTDI chip inside (FT2232H and/or FT4232H), on the other hand since you program a custom VID/PID for USB Blaster III, theoretically it should pick up your own Intel signed driver and not a standard FTDI driver, since it will not know what it is exactly. (Unless there is an overlap in some .sys files between your intel signed driver and the standard FTDI driver package.)

      For sure I will try it with a fresh computer and quartus installation too when we have more time, nevertheless there should be a way to make it work regardless of driver conflicts. Or at least it would be nice when it would fail gracefully and not with a BSOD.

      Kind regards,

      Peter

       

       

      • RichardT_altera's avatar
        RichardT_altera
        Icon for Super Contributor rankSuper Contributor
        1 month ago

        Does the BSOD still occur if you never power‑cycle the eval board?

        I’m asking because the behavior you’re seeing is consistent with a mixed‑FTDI driver environment, which is a plausible root cause.
        Even though the USB‑Blaster III uses an Intel‑signed INF with a custom VID/PID, both Intel’s package and the standard FTDI CDM packages typically load the same kernel‑level bus driver: ftdibus.sys. This SYS file is shared across many FTDI‑based devices.

        If Windows has multiple FTDI packages staged (e.g., one from Intel’s USB‑Blaster III and others for FT2232H/FT4232H products), the system can:

        -Select a different FTDI package on re‑enumeration (after a reboot or cable power cycle).
        -End up loading a different ftdibus.sys build (older or incompatible) for the same hardware instance.

        This aligns with your observation that a clean reinstall temporarily resolves the issue, but the crash returns later.
        Could you check whether multiple versions of ftdibus.sys exist on the machine?

        Regards,
        Richard Tan