From: Michal Pecio <michal.pecio@gmail.com>
To: The-Luga <lugathe2@gmail.com>
Cc: Alan Stern <stern@rowland.harvard.edu>,
Terry Junge <linuxsound@cosmicgizmosystems.com>,
linux-sound@vger.kernel.org, linux-usb@vger.kernel.org,
linux-input@vger.kernel.org
Subject: Re: [BUG] Edifier QR30 (2d99:a101, Jieli Technology) reboots itself when RGB brightness button is used under Linux
Date: Sun, 9 Nov 2025 09:24:50 +0100 [thread overview]
Message-ID: <20251109092450.693bcbe5.michal.pecio@gmail.com> (raw)
In-Reply-To: <CALvgqEAo8-MhE3ievoDkq4AOxRZ2E52kcko+GxYyf+WZE2H0=g@mail.gmail.com>
On Sun, 9 Nov 2025 02:18:44 -0300, The-Luga wrote:
> The speaker has a physical knob that changes brightness. When the
> speaker is disconnected from any USB port, the knob works fine
> increasing/decreasing brightness. When connected to Windows, rotating
> the knob (either increase or decrease) changes brightness normally and
> the speaker does not reboot (without needing any vendor software).
> When connected to Linux, rotating the knob causes the speaker to
> reboot.
>
> All brightness changes/reboots in the previous logs were done by
> rotating the speaker's physical knob.
>
> I believe this is a firmware bug on the speaker. My hypothesis: after
> each brightness change the speaker expects a particular USB response
> (the default response Windows sends). If it does not receive that
> response, the speaker thinks it is disconnected and resets.
I haven't attempted decoding your usbmon traces, but based on Terry's
comments, the device disconnects immediately when the knob is rotated
without even sending any HID packet. The original trace begins with
root hub traffic - likely disconnect notification, followed by failure
of a URB waiting for HID data from the speaker, and new enumeration.
If anything, it looks like Linux is doing something to the speaker
which makes it enter an invalid internal state and then the firmware
panics next time the knob is rotated, before anything is sent.
I wonder if there is some way to block 'usbhid' from ever binding to
this device (not just unbinding it later) and see if that hepls?
What happens if you pass to Windows and back, then rotate the knob?
Regards,
Michal
next prev parent reply other threads:[~2025-11-09 8:24 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <CALvgqEAq8ZWgG4Dyg_oL7_+nUDy+LUoTXi+-6aceO-AKtBS3Mg@mail.gmail.com>
2025-11-08 4:41 ` [BUG] Edifier QR30 (2d99:a101, Jieli Technology) reboots itself when RGB brightness button is used under Linux Terry Junge
2025-11-08 18:18 ` The-Luga
2025-11-08 20:48 ` Alan Stern
2025-11-09 0:15 ` The-Luga
2025-11-09 3:22 ` Alan Stern
2025-11-09 5:18 ` The-Luga
2025-11-09 8:24 ` Michal Pecio [this message]
2025-11-09 14:25 ` The-Luga
2025-11-09 15:32 ` Alan Stern
2025-11-09 16:44 ` The-Luga
2025-11-09 20:30 ` Alan Stern
2025-11-09 22:17 ` The-Luga
2025-11-09 22:49 ` Terry Junge
2025-11-10 0:56 ` The-Luga
2025-11-10 4:00 ` Terry Junge
2025-11-10 2:20 ` Alan Stern
2025-11-10 4:56 ` Dmitry Torokhov
2025-11-10 5:40 ` The-Luga
2025-11-10 6:54 ` The-Luga
2025-11-10 19:57 ` Terry Junge
2025-11-10 20:10 ` Alan Stern
2025-11-10 23:48 ` The-Luga
2025-11-11 1:59 ` The-Luga
2025-11-11 3:42 ` Alan Stern
2025-11-11 5:11 ` Terry Junge
2025-11-11 7:42 ` [PATCH] The Edifier QR30 USB speaker, identified as: Jieli Technology EDIFIER Hal0 2.0 SE 2d99:a101, reports a HID interface that needs HID_QUIRK_ALWAYS_POLL to ensure it does not crash when changing the RGB brightness with the physical knob Rodrigo Lugathe da Conceição Alves
2025-11-11 8:08 ` The-Luga
2025-11-11 19:33 ` Michal Pecio
2025-11-12 1:53 ` [PATCH v2] Apply the quirk HID_QUIRK_ALWAYS_POLL to the Edifier QR30 (2d99:a101) Rodrigo Lugathe da Conceição Alves
2025-11-12 5:20 ` Terry Junge
2025-11-12 17:25 ` Alan Stern
2025-11-13 15:45 ` The-Luga
2025-11-13 17:45 ` Terry Junge
2025-11-11 9:16 ` [BUG] Edifier QR30 (2d99:a101, Jieli Technology) reboots itself when RGB brightness button is used under Linux Oliver Neukum
2025-11-11 15:08 ` Alan Stern
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=20251109092450.693bcbe5.michal.pecio@gmail.com \
--to=michal.pecio@gmail.com \
--cc=linux-input@vger.kernel.org \
--cc=linux-sound@vger.kernel.org \
--cc=linux-usb@vger.kernel.org \
--cc=linuxsound@cosmicgizmosystems.com \
--cc=lugathe2@gmail.com \
--cc=stern@rowland.harvard.edu \
/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 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).