Skip to content

Est. 2011ยทMicrosoft Partner ID 7033487ยทDelivery under 3 minยทSupport 7 days a week

Your vault is empty.

Free Fix 0x00000144

BUGCODE_USB3_DRIVER 0x00000144 and USB crashes when a device is plugged in

11 min read Updated October 5, 2026 Windows Crashes & Boot Recovery

Fix it now

0x00000144 is the code used for all USB 3 bug checks, and parameter 1 says which one failed. That value decides everything: some point at the driver that owns the peripheral, some at the xHCI controller or its firmware, and some at a misbehaving hub. Read it before you start unplugging things.

The first in WinDbg with the dump open; the second in an elevated PowerShell session

!analyze -v
Get-PnpDevice -Class USB | Select-Object FriendlyName, Status
  1. Get parameter 1 from the dump or the blue screen. A value in the 0x1 to 0x3 or 0x800 range means a client driver misused the USB stack; 0x1001 upwards is the xHCI controller; 0x3000 upwards is the hub driver.
  2. If it is a client-driver value, work on the driver that owns the peripheral – the vendor’s driver for the dock, capture device, printer or audio interface – not on the peripheral itself.
  3. If it is 0x100C or 0x100F, the xHCI controller firmware is unsupported or outdated. That is a firmware update from the machine or board vendor.
  4. Install the chipset and USB host controller package from the motherboard or laptop vendor, then reboot.
  5. Only then isolate physically: unplug everything but keyboard and mouse, confirm stability, and reconnect one device at a time.

0x000000FE with parameter 1 of 0x3 is the code that genuinely supports a hardware reading – Microsoft says the USB miniport driver bug checking usually happens in response to a hardware failure. 0x00000144 is a different code with a different weighting.

If the machine is stable after the driver or firmware update, stop here. The next section lists what each band of parameter 1 values means.

Why it happens

The USB 3 stack is layered. The host controller driver talks to the xHCI silicon, a hub driver manages downstream ports, and class or vendor drivers handle individual functions. Each layer trusts the one below to return structurally valid data, and each layer can misuse the one below it. 0x00000144 is the single code used for all USB 3 bug checks, with parameter 1 identifying which internal check failed.

Read that parameter table and the picture is not the one most guides paint. The first band is client-driver misuse: 0x1 is a client driver using an URB it had previously sent to the core stack, 0x3 is a client driver sending a corrupted URB – either because it did not allocate the URB properly or because it caused a buffer underrun. The 0x800 band is static-streams and chained-MDL misuse, including opening streams before querying for the capability and sending an URB with a transfer buffer longer than the byte count of the MDL passed in.

0x2 is different and worth knowing: a boot or paging device failed re-enumeration. If your system disk is on USB, that is what you are looking at, and the peripheral hunt is beside the point.

The 0x1001 band is the xHCI controller: host system errors, host controller errors, failed stop and reset commands, and two firmware verdicts that are unusually actionable. 0x100C means the firmware in the xHCI controller is not supported and the driver will not load on that controller unless the firmware is updated; 0x100F means the firmware is outdated, the driver will continue but may run into issues, and an update is recommended. The 0x3000 band is the hub driver: a misbehaving hub reset successfully or unsuccessfully, a non-functional SuperSpeed hub disabled, and 0x3003, a USB device failing enumeration.

The driver that owns the peripheral is misusing the stack

You have this one if Parameter 1 in the 0x1 to 0x3 or 0x800 to 0x807 range.

  1. Update or remove the vendor driver for the device involved – dock, capture card, audio interface, industrial adapter. Class drivers are rarely the offender; vendor ones are.
  2. If the device works with the inbox class driver, use that instead of the vendor package as a test.
  3. These values describe a coding fault. If the current driver still does it, report it to the vendor rather than working around it.

The xHCI controller or its firmware

You have this one if Parameter 1 of 0x1001 upwards, and in particular 0x100C or 0x100F.

  1. 0x100C: the firmware is not supported and the xHCI driver will not load until it is updated. Get the firmware from the machine or board vendor.
  2. 0x100F: the firmware is outdated and an update is recommended. Same source.
  3. Install the chipset and USB host controller package from the same vendor rather than relying on the inbox driver.

A hub or a device that will not enumerate

You have this one if Parameter 1 in the 0x3000 to 0x3003 range, or crashes tied to one dock or hub.

  1. Move the device to a rear port on a desktop, connected directly to the board rather than through a front panel header or a hub.
  2. Replace unpowered hubs with powered ones where the attached devices draw real current.
  3. Update firmware for docks, hubs and external enclosures where the manufacturer publishes it – old bridge chip firmware is a frequent offender.

It is a different USB or PnP bug check

You have this one if 0x000000FE, 0x000000CA or 0x00000121 rather than 0x144.

  1. 0x000000FE BUGCODE_USB_DRIVER: an error in a USB driver. Parameter 1 of 0x3 is the USB miniport driver bug checking, which Microsoft says usually happens in response to a hardware failure.
  2. 0x000000CA PNP_DETECTED_FATAL_ERROR: the Plug and Play Manager hit a severe error, probably from a problematic PnP driver. Parameter 1 names the case – duplicate PDO, invalid PDO, invalid ID, invalid enumeration of a deleted PDO and others.
  3. 0x00000121 DRIVER_VIOLATION: parameter 1 of 0x1 means a driver called a function that can only be called at a specific IRQL; 0x2 means it called one at a greater IRQL than the function permits.

Full reference

Parameter 1 bands for 0x00000144

Range Layer Examples
0x1, 0x3 Client driver, URB misuse 0x1 reused an URB already sent to the core stack; 0x3 sent a corrupted URB
0x2 Boot or paging device A boot or paging device failed re-enumeration
0x800 – 0x805 Client driver, static streams Opening streams at raised IRQL, before querying capability, in an invalid number, or twice; failing to close a handle or free an URB
0x806 – 0x807 Client driver, chained MDLs Sending a chained MDL before querying the capability; a transfer buffer longer than the MDL byte count
0x1001 – 0x1022 xHCI controller Host system and host controller errors, failed stop, reset and configure commands, out-of-order completions
0x100C, 0x100F xHCI firmware Firmware not supported; firmware outdated
0x3000 – 0x3003 Hub driver Misbehaving hub reset or failed to reset; non-functional SuperSpeed hub disabled; a USB device failed enumeration

The size of the client-driver and controller bands is the point. Treating every 0x144 as “a bad peripheral” ignores most of the table, and sends you swapping cables while the actual fault sits in a vendor driver you never updated.

Bug checks that look like this one

Code Name Where the weight sits
0x00000144 BUGCODE_USB3_DRIVER All USB 3 bug checks. Parameter 1 decides which layer
0x000000FE BUGCODE_USB_DRIVER An error in a USB driver. Parameter 1 of 0x3 is the miniport bug checking, usually in response to a hardware failure
0x000000CA PNP_DETECTED_FATAL_ERROR The Plug and Play Manager, probably from a problematic PnP driver
0x00000121 DRIVER_VIOLATION A driver called a function at the wrong IRQL

0x000000FE’s parameter 1 values are worth a moment. 0x2 means a client driver submitted an URB still attached to another IRP pending in the bus driver; 0x4 means it submitted an IRP already pending; 0x5 is a hardware failure from a bad physical address in a hardware data structure; and 0x6 is a corrupted internal data structure. Only 0x3 and 0x5 point squarely at hardware.

0x000000CA, in more detail

  • 0x1, duplicate PDO: a driver enumerated multiple PDOs with identical device and unique IDs.
  • 0x2, invalid PDO: an API needing a PDO was called with random memory, an FDO, or an uninitialised PDO.
  • 0x3, invalid ID: an enumerator returned an ID containing illegal characters or not properly terminated.
  • 0x4, invalid enumeration of a deleted PDO: an enumerator returned a PDO it had previously deleted.
  • 0x5, PDO freed while linked in the devnode tree, usually a driver not adding a reference when returning the PDO in a query IRP.
  • 0x8, a NULL PDO returned as a bus relation.
  • All of them are driver faults in something that enumerates devices, which on a USB machine usually means a dock, a hub or a virtual bus driver.

Physical isolation, done properly

  1. Unplug every USB device except the keyboard and mouse, reboot, and confirm the machine is stable for long enough to mean something.
  2. Reconnect devices one at a time, using the machine for a few minutes after each. The last one added when the crash returns is the suspect.
  3. Move the suspect to a different port – rear rather than front, directly on the board rather than through a hub.
  4. Test the same device on another machine. A device that crashes two machines is faulty; one that crashes only this one points at this machine’s driver or firmware.
  5. Get-PnpDevice -Class USB | Select-Object FriendlyName, Status gives you the current inventory and any device reporting an error state.

When it crashes with nothing attached

  • Update the system firmware. The xHCI firmware verdicts, 0x100C and 0x100F, are explicit that this can be the whole problem.
  • Install the chipset and USB package from the machine vendor rather than relying on what Windows Update provided.
  • Check the System log around each crash for device errors that precede the bug check.
  • Test the keyboard in a different port before suspecting the controller itself – internal card readers, fingerprint sensors and webcams are USB devices too, and they are attached whether or not anything is plugged in externally.

Every code this article covers

Code What it points at Source
0x00000144 BUGCODE_USB3_DRIVER: the code used for all USB 3 bug checks. Parameter 1 specifies which check failed – client driver URB misuse, boot or paging device re-enumeration, static streams and chained MDLs, xHCI controller errors, and hub driver errors Microsoft Learn
0x000000FE BUGCODE_USB_DRIVER: an error occurred in a USB driver. Parameter 1 of 0x3 is the USB miniport driver bug checking, which Microsoft says usually happens in response to a hardware failure Microsoft Learn
0x000000CA PNP_DETECTED_FATAL_ERROR: the Plug and Play Manager encountered a severe error, probably as a result of a problematic Plug and Play driver. Parameter 1 names the case, such as a duplicate PDO or an invalid device ID Microsoft Learn
0x00000121 DRIVER_VIOLATION: a driver called a function that can only be called at a specific IRQL (parameter 1 of 0x1), or called a function at a greater IRQL than the function permits (0x2) Microsoft Learn

Confirm the fix worked

  1. The machine runs with the full set of devices attached, through the same ports, without a new dump.
  2. Get-PnpDevice -Class USB shows every device with an OK status rather than an error.
  3. The suspect device works on another machine, if you concluded the fault was this machine’s driver or firmware.
  4. The System log records no USB device errors over the tested period.

Questions people ask about this

Is 0x144 always a faulty peripheral?

No, and the parameter table is the reason. Large parts of it describe client drivers misusing the USB stack and xHCI controller or firmware errors. Read parameter 1 before assuming a cable or a hub is at fault – it will often send you to a driver or a firmware update instead.

Which code actually points at hardware?

0x000000FE with parameter 1 of 0x3. Microsoft says the USB miniport driver bug checking usually happens in response to a hardware failure. 0x5 in the same code is also hardware – a bad physical address in a hardware data structure.

The dump gives parameter 1 of 0x100C. What does that mean?

The firmware in the xHCI controller is not supported, and the xHCI driver will not load on that controller until the firmware is updated. Get the firmware from the machine or board vendor. 0x100F is the softer version: outdated firmware, driver still works, update recommended.

Should I disable USB selective suspend?

It is a reasonable test on a machine that crashes at idle, but it is not the documented fix and it costs battery life if left off. Work from parameter 1 first, because most of the values in that table are not power-related at all.

Related error codes

Was this article helpful?

Your feedback helps us improve our documentation.

Related articles

License Error END_OF_NT_EVALUATION_PERIOD 0x00000098 – evaluation Windows Server shuts down Free Fix UNEXPECTED_STORE_EXCEPTION 0x00000154 on Windows 10 and Windows 11 Free Fix VIDEO_TDR_FAILURE 0x00000116 – display driver stopped responding Free Fix MANUALLY_INITIATED_CRASH 0x000000E2 and missing memory dump files
โ† Back to Knowledge Base