From: "Michał Pecio" <michal.pecio@gmail.com>
To: linux-usb@vger.kernel.org
Subject: "Transfer event TRB DMA ptr not part of current TD" spam after USB disconnection
Date: Fri, 12 Jan 2024 23:52:05 +0100 [thread overview]
Message-ID: <20240112235205.1259f60c@foxbook> (raw)
Hi,
This is a long standing issue seen in many kernel versions up to 6.7.
Disconnecting a high- or super-speed UVC camera from USB 3.0 port during
recording prints a lot of these errors (usually much more than here).
This appears to come from handle_tx_event() and it is my understanding
that the function tries to look up a transfer request previously issued
by the driver which is pointed to by a completion event received from
the hardware, and the pointer is found to be out of expected bounds.
It is a consistent pattern that problematic event-dma pointers are
within segment bounds, but they start one slot past trb-end and grow.
Comments in the code suggest that "the HC is hosed", but I suspect the
real problem might be the driver failing to retire the _previous_ TD
after it completed with similar errors and continuing to wait for its
completion events forever, or something else with similar outcome.
Or, at any rate, I *hope* that it's a harmless bug and not some memory
corruption... I only have very basic understanding of XHCI.
Any thoughts or suggestions how to fix or debug this further?
Thanks,
Michal
# camera connected
[ 2968.873423] usb 8-1: new high-speed USB device number 54 using xhci_hcd
[ 2969.155383] usb 8-1: New USB device found, idVendor=0bda, idProduct=58b0, bcdDevice= 7.01
[ 2969.155393] usb 8-1: New USB device strings: Mfr=3, Product=1, SerialNumber=2
[ 2969.155396] usb 8-1: Product: FULL HD 1080P Webcam
[ 2969.155398] usb 8-1: Manufacturer: Generic
[ 2969.155400] usb 8-1: SerialNumber: 200901010001
[ 2969.162076] usb 8-1: Found UVC 1.00 device FULL HD 1080P Webcam (0bda:58b0)
# recording started
[ 2978.289442] usb 8-1: Device requested 3072 B/frame bandwidth
[ 2978.289461] usb 8-1: Selecting alternate setting 7 (3072 B/frame bandwidth)
[ 2978.291230] usb 8-1: Allocated 5 URB buffers of 32x3072 bytes each
[ 2978.910333] usb 8-1: Device requested 3072 B/frame bandwidth
[ 2978.910352] usb 8-1: Selecting alternate setting 7 (3072 B/frame bandwidth)
[ 2978.913549] usb 8-1: Allocated 5 URB buffers of 32x3072 bytes each
# camera disconnected
[ 2988.728338] xhci_hcd 0000:02:00.0: ERROR Transfer event TRB DMA ptr not part of current TD ep_index 2 comp_code 4
[ 2988.728348] xhci_hcd 0000:02:00.0: Looking for event-dma 000000012d313890 trb-start 000000012d313870 trb-end 000000012d313880 seg-start 000000012d313000 seg-end 000000012d313ff0
[ 2988.728449] xhci_hcd 0000:02:00.0: ERROR Transfer event TRB DMA ptr not part of current TD ep_index 2 comp_code 4
[ 2988.728452] xhci_hcd 0000:02:00.0: Looking for event-dma 000000012d3138a0 trb-start 000000012d313870 trb-end 000000012d313880 seg-start 000000012d313000 seg-end 000000012d313ff0
[ 2988.728527] usb 8-1: USB disconnect, device number 54
[ 2988.728573] xhci_hcd 0000:02:00.0: ERROR Transfer event TRB DMA ptr not part of current TD ep_index 2 comp_code 4
[ 2988.728576] xhci_hcd 0000:02:00.0: Looking for event-dma 000000012d3138b0 trb-start 000000012d313870 trb-end 000000012d313880 seg-start 000000012d313000 seg-end 000000012d313ff0
[ 2988.728698] xhci_hcd 0000:02:00.0: ERROR Transfer event TRB DMA ptr not part of current TD ep_index 2 comp_code 4
[ 2988.728701] xhci_hcd 0000:02:00.0: Looking for event-dma 000000012d3138c0 trb-start 000000012d313870 trb-end 000000012d313880 seg-start 000000012d313000 seg-end 000000012d313ff0
[ 2988.728841] xhci_hcd 0000:02:00.0: ERROR Transfer event TRB DMA ptr not part of current TD ep_index 2 comp_code 4
[ 2988.728844] xhci_hcd 0000:02:00.0: Looking for event-dma 000000012d3138d0 trb-start 000000012d313870 trb-end 000000012d313880 seg-start 000000012d313000 seg-end 000000012d313ff0
[ 2988.728948] xhci_hcd 0000:02:00.0: ERROR Transfer event TRB DMA ptr not part of current TD ep_index 2 comp_code 4
[ 2988.728951] xhci_hcd 0000:02:00.0: Looking for event-dma 000000012d3138e0 trb-start 000000012d313870 trb-end 000000012d313880 seg-start 000000012d313000 seg-end 000000012d313ff0
[ 2988.729073] xhci_hcd 0000:02:00.0: ERROR Transfer event TRB DMA ptr not part of current TD ep_index 2 comp_code 4
[ 2988.729075] xhci_hcd 0000:02:00.0: Looking for event-dma 000000012d3138f0 trb-start 000000012d313870 trb-end 000000012d313880 seg-start 000000012d313000 seg-end 000000012d313ff0
[ 2988.729197] xhci_hcd 0000:02:00.0: ERROR Transfer event TRB DMA ptr not part of current TD ep_index 2 comp_code 4
[ 2988.729199] xhci_hcd 0000:02:00.0: Looking for event-dma 000000012d313900 trb-start 000000012d313870 trb-end 000000012d313880 seg-start 000000012d313000 seg-end 000000012d313ff0
[ 2988.729322] xhci_hcd 0000:02:00.0: ERROR Transfer event TRB DMA ptr not part of current TD ep_index 2 comp_code 4
[ 2988.729324] xhci_hcd 0000:02:00.0: Looking for event-dma 000000012d313910 trb-start 000000012d313870 trb-end 000000012d313880 seg-start 000000012d313000 seg-end 000000012d313ff0
[ 2988.729447] xhci_hcd 0000:02:00.0: ERROR Transfer event TRB DMA ptr not part of current TD ep_index 2 comp_code 4
[ 2988.729450] xhci_hcd 0000:02:00.0: Looking for event-dma 000000012d313920 trb-start 000000012d313870 trb-end 000000012d313880 seg-start 000000012d313000 seg-end 000000012d313ff0
[ 2988.729572] xhci_hcd 0000:02:00.0: ERROR Transfer event TRB DMA ptr not part of current TD ep_index 2 comp_code 4
[ 2988.729575] xhci_hcd 0000:02:00.0: Looking for event-dma 000000012d313930 trb-start 000000012d313870 trb-end 000000012d313880 seg-start 000000012d313000 seg-end 000000012d313ff0
[ 2988.729697] xhci_hcd 0000:02:00.0: ERROR Transfer event TRB DMA ptr not part of current TD ep_index 2 comp_code 4
[ 2988.729700] xhci_hcd 0000:02:00.0: Looking for event-dma 000000012d313940 trb-start 000000012d313870 trb-end 000000012d313880 seg-start 000000012d313000 seg-end 000000012d313ff0
next reply other threads:[~2024-01-12 22:52 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-01-12 22:52 Michał Pecio [this message]
2024-01-13 20:47 ` "Transfer event TRB DMA ptr not part of current TD" spam after USB disconnection Michał Pecio
2024-01-14 14:06 ` Michał Pecio
2024-01-15 13:58 ` Mathias Nyman
2024-01-15 16:27 ` Michał Pecio
2024-01-16 15:36 ` [RFT PATCH] xhci: process isoc TD properly when there was an error mid TD Mathias Nyman
2024-01-16 22:20 ` Michał Pecio
2024-01-17 10:46 ` Mathias Nyman
2024-01-17 17:49 ` Michał Pecio
2024-01-18 11:00 ` Isochronous error handling bug on VIA VL805 Michał Pecio
2024-01-18 11:10 ` Michał Pecio
2024-01-18 13:54 ` Mathias Nyman
2024-01-18 13:56 ` [RFT PATCH v2] xhci: process isoc TD properly when there was an error mid TD Mathias Nyman
2024-01-18 22:16 ` Michał Pecio
2024-01-19 10:49 ` Mathias Nyman
2024-01-19 10:58 ` [RTF PATCH v3] " Mathias Nyman
2024-01-19 21:54 ` Michał Pecio
2024-01-22 9:03 ` Michał Pecio
2024-01-22 13:37 ` Mathias Nyman
2024-01-22 17:10 ` Michał Pecio
2024-01-22 11:47 ` Mathias Nyman
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20240112235205.1259f60c@foxbook \
--to=michal.pecio@gmail.com \
--cc=linux-usb@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.