Re: [BUG] thunderbolt: USB4 v2 TBGAA tunnel creation crash in TMU enhanced uni-directional mode

[Mika Westerberg <mika.westerberg@linux.intel.com>]

Hi,

On Tue, May 26, 2026 at 06:03:39PM +0200, Sebastian Loscher wrote:
> Hey all,
> 
> Thanks for your reply. I finally managed to attach the tbtrace logfile and
> dmesg, done two times after fresh reboot: One set after connecting the dock to
> the designated USB4V2 labelled (back) ports - reconnected three times, one set
> to the standard usbc ports in the front (labelled -amd).
> With the latter, dock stays attached and shares onboard LAN chip, SSD port and
> additional usb ports.
> dmesg shows same behavior as reported before on the usb4v2 ports.
> 
> I believe vendor added the usb4v2 on some pci lanes where other vendors went an
> occulink connector, additional M.2 and/or more USB ports.
> 
> Hope this helps. Let me know if you need more.

Thanks for the logs! I can see that there is unplug in the middle of USB 3
path setup here:

[  148.157866] tb_rx Read Response Domain 0 Route 1 Adapter 1 
               0x00/---- 0x80000000 0b10000000 00000000 00000000 00000000 .... Route String High
               0x01/---- 0x00000001 0b00000000 00000000 00000000 00000001 .... Route String Low
               0x02/---- 0x020820c1 0b00000010 00001000 00100000 11000001 .... 
                 [00:12]       0xc1 Address
                 [13:18]        0x1 Read Size
                 [19:24]        0x1 Adapter Num
                 [25:26]        0x1 Configuration Space (CS) → Adapter Configuration Space
                 [27:28]        0x0 Sequence Number (SN)
               0x03/00c1 0x81120408 0b10000001 00010010 00000100 00001000 .... 
[  148.159120] tb_tx Read Request Domain 0 Route 1 Adapter 1 
               0x00/---- 0x00000000 0b00000000 00000000 00000000 00000000 .... Route String High
               0x01/---- 0x00000001 0b00000000 00000000 00000000 00000001 .... Route String Low
               0x02/---- 0x020820c1 0b00000010 00001000 00100000 11000001 .... 
                 [00:12]       0xc1 Address
                 [13:18]        0x1 Read Size
                 [19:24]        0x1 Adapter Num
                 [25:26]        0x1 Configuration Space (CS) → Adapter Configuration Space
                 [27:28]        0x0 Sequence Number (SN)
[  148.160517] tb_event Hot Plug Event Packet Domain 0 Route 0 Adapter 1 / Lane
               0x00/---- 0x80000000 0b10000000 00000000 00000000 00000000 .... Route String High
               0x01/---- 0x00000000 0b00000000 00000000 00000000 00000000 .... Route String Low
               0x02/---- 0x80000001 0b10000000 00000000 00000000 00000001 .... 
                 [00:05]        0x1 Adapter Num
                 [31:31]        0x1 UPG
[  148.160518] tb_tx Notification Packet Domain 0 Route 0 
               0x00/---- 0x00000000 0b00000000 00000000 00000000 00000000 .... Route String High
               0x01/---- 0x00000000 0b00000000 00000000 00000000 00000000 .... Route String Low
               0x02/---- 0xc0000107 0b11000000 00000000 00000001 00000111 .... 
                 [00:07]        0x7 Event Code → HP_ACK
                 [08:13]        0x1 Event Info
                 [14:14]        0x0 Sequence
                 [30:31]        0x3 PG
[  148.169868] usb 5-1.4: USB disconnect, device number 5

There is also USB 2.x disconnect which makes me suspect the cable. Are you
using Thunderbolt 4/5 cable or something else?

The other thing I noticed is that the host is USB4v2 as well (well
according to the TMU configuration). @Mario is that right? I was not sure
if AMD is v2.