On 7/11/21 12:38 PM, Greg KH wrote:
On Sun, Jul 11, 2021 at 10:25:03AM +0200, Tristan Miller wrote:
Greetings.
Many models of Logitech USB webcams malfunction under Linux by producing
audio that sounds distorted, garbled, or high-pitched. For most users,
the problem happens randomly, and can usually be worked around by
restarting the audio application or by unplugging the webcam and then
plugging it back in again.
The problem is described in a kernel.org bug at
<https://bugzilla.kernel.org/show_bug.cgi?id=44281>. Despite eight
years' worth of comments and duplicate bugs indicating that that the
problem is still occurring, this report has remained closed as
"resolved" since 2013. It's unclear whether any of the right developers
have even seen these comments; certainly none of them have ever
responded, even just to reopen the bug. A comment on one of the
duplicate bugs <https://bugzilla.kernel.org/show_bug.cgi?id=203763#c1>
suggests that this may be because the bug was misfiled in the
v4l-dvb/webcam product, whereas it should instead have been posted to
this mailing list.
I'd appreciate it if one of the USB for Linux developers could take a
quick look at the bug report to confirm whether it's been filed in the
right place. If not, please let us know (either here or in Bugzilla)
where the bug should be reported instead, and if it is indeed here on
linux-usb, let us know if any further diagnostic information is
required.
Please report the information directly here, to this list, and to the
sound developers at alsa-devel@xxxxxxxxxxxxxxxx and we can all work to
resolve this.
First off it would be great to test if this happens in 5.13, does it?
And how can you reproduce it and with what specific device?
Hi,
I just recently bought a very expensive USB 3.0 Logitech BRIO 4K USB
webcam. These webcams provide a standard USB audio compliant interface,
and thanks for that! I've noticed that when doing a looptest, I.E.
playing the recorded audio back through a set a of headphones, then
after some 30 minutes, the audio starts corrupting. This does not happen
on MacOSX. I've dumped the audio data using both Ellisys USB2.0 and
usbdump on FreeBSD, and verified that it is not a host software problem.
The data stream from the device really carries the bad audio. Now just
recently someone reported to me a problem with a "Scarlett Solo 3rd"
under FreeBSD with XHCI, and I started looking into the code and found
some bugs* regarding the start frame for isochronous packets. After that
was fixed it seems to work fine again, but I don't know for how long.
I suspect these Logitech devices have received very little so-called
negative testing. I was in contact with Logitech support, but the people
there know basically nothing, and it ended they offered me a warranty
refund - how funny!
So yes, there are some software problems in these devices the vendor is
not willing to acknowledge, and they can suddenly trigger.
I know that FreeBSD is not Linux, but I thought that sharing my
experiences here might help moving forward with this issue!
--HPS
*)
https://cgit.freebsd.org/src/commit/?id=d038463bd269031fa8f13c0951b223bcae02bbca
https://cgit.freebsd.org/src/commit/?id=8fc2a3c41791b205a107dc2bec16ac7514a57958
--HPS
