Re: Regression: webcam freezing since Linux 6.15

[Michal Pecio <michal.pecio@gmail.com>]

On Thu, 23 Jul 2026 15:45:25 -0700, Bart Nagel wrote:
> I found that I can reproduce the freeze much, much faster if I run
> `stress --cpu $(nproc)` at the same time.

Great news, good work.

What happens differently in absence of CPU load?
- no Missed Service Errors anymore
- ep_trb_dma (see below) becomes always zero
- no obvious change, somehow the kernel gets more lucky

>     [  174.709692] xhci_hcd 0000:00:14.0: Transfer error for slot 6 ep 2 on endpoint
>     [  174.709815] xhci_hcd 0000:00:14.0: Miss service interval error for slot 6 ep 2 ep_trb_dma 0, set skip flag

That's kinda normal. Errors sometimes happen, isochronous doesn't
recover from them, SW must deal with it. Transfer error becomes -71
status, Missed Service becomes -18. Later uvcvideo logs this again.

ep_trb_dma should point to the failed transfer, but xHCI 1.0 spec
allows it to be zero for Missed Service Error. We deal with that too.

But oddly, on your HW it sometimes isn't zero. The commit you bisected
assumes that non-zero numbers are valid pointers and this seems to work
more or less correctly, until it doesn't:

[  174.818444] xhci_hcd 0000:00:14.0: Miss service interval error for slot 6 ep 2 ep_trb_dma 108165b90, set skip flag
[  174.818452] xhci_hcd 0000:00:14.0: Found td. Clear skip flag for slot 6 ep 2.
--
[  176.818812] xhci_hcd 0000:00:14.0: Miss service interval error for slot 6 ep 2 ep_trb_dma 108165dd0, set skip flag
[  176.818817] xhci_hcd 0000:00:14.0: Found td. Clear skip flag for slot 6 ep 2.
--
[  177.517933] xhci_hcd 0000:00:14.0: Miss service interval error for slot 6 ep 2 ep_trb_dma 10a63ec30, set skip flag
[  177.517937] xhci_hcd 0000:00:14.0: Found td. Clear skip flag for slot 6 ep 2.
--
[  177.586809] xhci_hcd 0000:00:14.0: Miss service interval error for slot 6 ep 2 ep_trb_dma 10a63e020, set skip flag
[  177.586815] xhci_hcd 0000:00:14.0: Found td. Clear skip flag for slot 6 ep 2.
--
[  177.636060] xhci_hcd 0000:00:14.0: Miss service interval error for slot 6 ep 2 ep_trb_dma 10a63e530, set skip flag
[  177.636065] xhci_hcd 0000:00:14.0: Found td. Clear skip flag for slot 6 ep 2.
--
[  177.817933] xhci_hcd 0000:00:14.0: Miss service interval error for slot 6 ep 2 ep_trb_dma 10a63ee10, set skip flag
[  177.817937] xhci_hcd 0000:00:14.0: Found td. Clear skip flag for slot 6 ep 2.
--
[  179.934449] xhci_hcd 0000:00:14.0: Miss service interval error for slot 6 ep 2 ep_trb_dma 10a63e7b0, set skip flag
[  179.934453] xhci_hcd 0000:00:14.0: Found td. Clear skip flag for slot 6 ep 2.
--
[  180.701300] xhci_hcd 0000:00:14.0: Miss service interval error for slot 6 ep 2 ep_trb_dma 10a63e920, set skip flag
[  180.701315] xhci_hcd 0000:00:14.0: All TDs skipped for slot 6 ep 2. Clear skip flag.

For some reason the driver thought that all TDs were missed and
returned them with -18 status; we see a burst of those logs from
uvcvideo moment later. Then I suspect uvcvideo resubmitted enough
URBs to go around the whole ring and overwrite pending transfers
with opposite cycle bit, which caused the HW to stop and produce
a Ring Overrun event.

The root cause is bogus ep_trb_dma from the HW or some kernel bug.

Maybe we could get enough info from debugfs to see what happened. If
possible, disconnect other USB devices from bus 1 and 2, trigger the
bug again, and *before* stopping video recording, run:

zip -r debugfs.zip /sys/kernel/debug/usb/xhci/0000:00:14.0

This requires debugfs to be enabled in the kernel and mounted. And zip
will print some warnings, but generate a valid archive. If you need to
use tar instead, you must first copy the directory, then archive it.

Email the zip (should be under 1MB) and corresponding dmesg as before.

Regards,
Michal