WSL2 Mirrored networking mode reproducibly triggers 0x9F DRIVER_POWER_STATE_FAILURE (Intel BE200 Wi-Fi) at every boot

Open
#42,039 3 comments 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
5/5
Estimated time
Over a week
Newbie friendliness
25/100
Issue type
Bug
Clarity
Mostly clear
Activity status
Active
Tech stack
cpp, linux
Domain
networking

Research direction

Start by reviewing the attached minidump and event log alongside the reported kernel dump stack, focusing on the blocked power IRP in the Wi-Fi miniport stop path. Compare boot behavior with Mirrored and NAT networking modes; done means Mirrored mode boots without the 0x9F bugcheck or hard hang on the reported Intel BE200 setup.

Written by the indexing model from the issue text.

Description

network wsl2
Windows Version

Windows 11 Pro, build 26100 (24H2)

WSL Version

WSL: 3.0.3.0

Are you using WSL 1, WSL 2, or WSLC?
  • WSL 2
  • WSL 1
  • WSLC
Kernel Version

6.18.54.1-1

Distro Version

Ubuntu-22.04

Other Software

None. The issue reproduces with WSL alone - no Docker, VS Code, or third-party software involved.

Repro Steps

rerequisite: Intel Wi-Fi 7 BE200 wireless adapter (M.2 CNVi), WSL 2.

  1. Set %USERPROFILE%.wslconfig to:

    [wsl2]
    networkingMode=Mirrored
    firewall=false

  2. Run wsl --shutdown, then reboot the machine.

  3. No user action needed after reboot. The failure occurs 8-13 seconds after boot,
    immediately after the Wi-Fi driver logs "Wi-Fi Driver Enabled"
    (Netwaw18/Netwaw20 event 7010 in the System log).

Result: 100% reproducible - every single boot with Mirrored mode.

Control test (same machine, same steps, only networkingMode changed):

  • networkingMode=NAT -> boots cleanly every time, zero failures
Expected Behavior

The system should boot normally with WSL Mirrored networking mode enabled.
The Hyper-V virtual switch should bind to the physical Wi-Fi adapter without
destabilizing the host, and WSL should start and function normally - exactly
as it does with networkingMode=NAT.

Actual Behavior

Two failure modes, both requiring a forced power-off (long-press power button)
to recover - a normal shutdown hangs:

Mode A - Bugcheck:
BugCheck 0x0000009F DRIVER_POWER_STATE_FAILURE
Arg1: 0x0000000000000003 (a device object has been blocking an IRP too long)
Arg2: 0xffff820a0474e060 (Physical Device Object of the stack)
Arg3: 0xffffe1874bd8f5d0 (nt!TRIAGE_9F_POWER)
Arg4: 0xffff820a198b9050 (the blocked IRP)
MEMORY.DMP is written.

Mode B - Hard hang:
No bugcheck, no dump. System becomes unresponsive; Event 6008 records
"the previous system shutdown ... was unexpected".

After forced reboot the adapter recovers, usually after a period of degraded
connectivity (WLAN-AutoConfig event 4003 "limited connectivity").

Event-log timeline from a crashed boot (Mirrored mode):

15:02:33  Kernel-Boot events (system start)
15:02:36  VfpExt / Hyper-V-VmSwitch burst: 40 events   <- mirrored-mode switch binding
15:02:38  Netwaw18 7010 "Wi-Fi Driver Enabled"
15:02:46  CRASH  (13 s after boot)
15:02:48  MEMORY.DMP written; WER 1001 reports 0x9F
15:02:50  VfpExt / Hyper-V-VmSwitch burst: 22 events   <- WSL auto-start
15:03:23  WLAN-AutoConfig 4003 "limited connectivity"
15:03:24  Netwaw18 8000 "Miniport Halt" + re-initialization

Kernel dump stack - the power IRP is blocked in the miniport stop path:

nt!PopIrpWorker
  Wdf01000!FxPkgFdo::DispatchDeviceSetPower
    Wdf01000!FxPkgPnp::PowerGotoDx
      WifiCx!EvtCxDevicePostSelfManagedIoSuspendEx
        -> DeletePortTypes -> NetAdapterStop
          ndis!ndisPnPStopDevice / ndisPnPRemoveDevice
            tcpip!FlpWaitForMiniportToReturnTransmittedPackets   <- blocked indefinitely
Diagnostic Logs

KEY FINDING 1 - Controlled experiment (only variable is networkingMode):
Mirrored : boot-time VfpExt/VmSwitch burst = 40 events, overlapping BE200
driver initialization -> crash at EVERY boot
NAT : boot-time burst = 2 events -> completely stable
(no crash, no miniport halt, no 4003 across multiple reboots,
sleep/wake cycles and WSL start/stop cycles)

KEY FINDING 2 - Not driver-version specific:
Reproduced on both 24.70.0.3 (Netwaw18.sys) and 24.80.0.2 (Netwaw20.sys).

KEY FINDING 3 - Not hardware:
HP Support Assistant wireless diagnostic = PASS; the adapter works fine
under NAT mode. The failure is deterministic and purely software-triggered.

Workaround: networkingMode=NAT is completely stable.

Attached files:
- Kernel minidump from a crash (9.8 MB)
- System event log export (CSV) covering multiple crash timestamps

Root-cause analysis: Mirrored mode binds the Hyper-V extensible switch to the
PHYSICAL Wi-Fi adapter during boot, forcing the BE200 miniport through a
stop/remove cycle that races with the adapter's own initialization. The driver
never returns outstanding TX packets, so the power IRP never completes.

Requests:

  1. WSL team: can the vSwitch binding to physical adapters be deferred until
    the host network stack is fully initialized at boot, instead of racing
    with adapter initialization?
  2. Intel: Netwaw*.sys should complete power IRPs within the timeout even
    with in-flight TX packets.

minidump_102020.dmp

hp_evidence_events.csv

Dominant language
C++
Stars
34.4k
Forks
1.9k
Avg merge
1d 23h
Merged PRs (30d)
166

Getting set up

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from microsoft/WSL

All issues in microsoft/WSL

Similar issues

More C++ issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.