## Update: undocked-boot and USB-C port control tests on 7.0.0-28

  I performed a controlled reboot on 2026-07-21 with the dock
disconnected. I waited until the graphical desktop was running before
connecting the same dock, cable, two DisplayPort monitors, and
peripherals.

  The undocked boot did not produce the previously observed boot-time
messages:

  - `failed to allocate DP resource for port 7`
  - `DMUB notification skipped due to no handler`
  - `DM_MST: Differing MST start`

  However, the post-login result depended on which laptop USB-C port was
used.

  ### Laptop USB-C port 1 → dock rear/upstream port: failed

  At 09:35:56, the connection failed during the initial USB setup,
before the dock, Thunderbolt topology, or displays could enumerate:

  ```text
  usb 8-1: new high-speed USB device number 2 using xhci_hcd
  usb 8-1: Device not responding to setup address.
  usb 8-1: Device not responding to setup address.
  usb 8-1: device not accepting address 2, error -71

  No subsequent Thunderbolt, USB4, or DP MST topology appeared.
Disconnecting the cable produced no device-removal event, consistent
with the device never completing enumeration.

  ### Laptop USB-C port 2 → dock front USB-C port: no upstream event

  This produced no USB, USB4, Thunderbolt, or DRM hot-plug event. All
external DP connectors remained disconnected. This is consistent with
the dock’s front USB-C port being downstream/peripheral-facing rather
than its host/upstream port.

  ### Laptop USB-C port 2 → dock rear/upstream port: succeeded

  Using the same dock, cable, displays, peripherals, running kernel, and
desktop session, the dock enumerated through the second laptop USB-C
port:

  [drm] DM_MST: starting TM
  [drm] DM_MST: DP14, 2-lane link detected
  usb 7-1: new SuperSpeed Plus Gen 2x1 USB device number 2
  usb 7-1: New USB device found, idVendor=8087, idProduct=0b40
  usb 7-1: Product: USB3.0 Hub

  The dock MCU, USB 2/3 hubs, Realtek Ethernet, dock audio, and attached
Logitech C270 webcam all enumerated. Ethernet obtained carrier.

  Display topology discovery logged one error:

  amdgpu 0000:33:00.0: [drm] *ERROR* Sending link address failed with -5

  The driver recovered without disconnecting the dock:

  [drm] Synaptics Cascaded MST hub
  [drm] Synaptics Cascaded MST hub
  [drm] MST_DSC Send DSC enable to synaptics
  [drm] MST_DSC Send DSC enable to synaptics

  Both external screens then turned on. The final DRM connector state
showed DP-9 and DP-10 connected.

  This successful attempt did not produce:

  - USB error -71
  - dcn31_program_compbuf_size warning
  - update_dpia_stream_allocation_table warning
  - GNOME/Mutter atomic-commit failure
  - complete dock disconnect/reconnect

  ## Interpretation

  These results do not support a purely boot-ordering or purely AMDGPU
explanation.

  Booting undocked avoids the previously observed boot-time DP-
resource/MST-ordering messages, but a normal post-login connection can
still fail at the earliest USB setup stage on one laptop USB-C path. At
that point, Thunderbolt and AMDGPU have not yet discovered the dock.

  The same equipment works minutes later through the second laptop USB-C
port, although that successful path still produces a transient
recoverable AMDGPU MST link-address error.

  The normal-use failure therefore appears port/path-sensitive and may involve 
the platform USB/XHCI/USB4 path, firmware, signal integrity, or hardware. The 
available evidence does not yet identify which component is responsible. 
Separately, the trigger_hpd_mst_set kernel oops remains a definite AMDGPU 
kernel-driver
  defect.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2160619

Title:
  USB-C/USB4 dock path instability: port-dependent USB error -71,
  Thunderbolt disconnects, and AMDGPU DP MST failures

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2160619/+subscriptions


-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to