LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Juergen Stuber <juergen@jstuber.net>
To: linux-kernel@vger.kernel.org
Cc: Ricky Wu <ricky_wu@realtek.com>,
	Mathias Nyman <mathias.nyman@linux.intel.com>
Subject: Thunderbolt 3 Dock, USB issues when undocking then docking
Date: Tue, 7 Sep 2021 20:23:10 +0200	[thread overview]
Message-ID: <20210907202310.77dea423@sju.home.jstuber.net> (raw)

Greetings!

In Linux versions 5.10-rc1 or newer (exact bisect see below) I have
issues when I undock and then dock my Thunderbolt 3 dock.
Afterwards I see

- log messages like "retire_capture_urb: 571 callbacks suppressed"
  (always)
- audio glitches when playing audio via the dock on USB speakers
  (almost always)
- keyboard events are lost on a dock-connected keyboard (sometimes)

I tested this by undocking, waiting a short while, then redocking, and
after that playing audio via USB speakers connected to the dock.
It seems to be deterministic, that is, in affected kernels it seems to
happen on every undock-dock cycle.

This is a Thinkpad T14 Gen1 Intel (20S0000HGE) and
a Thinkpad Thunderbolt 3 Dock Gen2 (40AN / SD20M70247), 
the USB speakers are Logitech Z-10,
the distribution is Debian bullseye.

I bisected the issue, this pointed to the first bad commit

726eb70e0d34 Merge tag 'char-misc-5.10-rc1' of
git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/char-misc

Since that was a big merge and thus not too helpful, I bisected the two
branches that were merged. That is, I bisected one branch and merged
into the head of the other branch and tested the merge.
On char-misc this pointed to

7c33e3c4c79a misc: rtsx: Add power saving functions and fix driving
parameter

The affected driver is in use for a device in my system:
02:00.0 Unassigned class [ff00]: Realtek Semiconductor Co., Ltd.
RTS522A PCI Express Card Reader [10ec:522a] (rev 01)

Log output:
[    1.179205] rtsx_pci 0000:02:00.0: enabling device (0000 -> 0002)

On the mainline side the bisect points to

edc649a82341 xhci: Tune interrupt blocking for isochronous transfers


I tried to revert each of these, but that was not easily possible.


RedHat has a bug report about this issue:
https://bugzilla.redhat.com/show_bug.cgi?id=1951372

Please ask if you need more info or testing.


Jürgen

-- 
Jürgen Stuber <juergen@jstuber.net>
http://www.jstuber.net/


                 reply	other threads:[~2021-09-07 23:29 UTC|newest]

Thread overview: [no followups] expand[flat|nested]  mbox.gz  Atom feed

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=20210907202310.77dea423@sju.home.jstuber.net \
    --to=juergen@jstuber.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mathias.nyman@linux.intel.com \
    --cc=ricky_wu@realtek.com \
    --subject='Re: Thunderbolt 3 Dock, USB issues when undocking then docking' \
    /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

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).