Linux Sound subsystem development
 help / color / mirror / Atom feed
From: Takashi Iwai <tiwai@suse.de>
To: Cristian-Marian Fotache <cristianmarianfotache@gmail.com>
Cc: tiwai@suse.com, perex@perex.cz, linux-sound@vger.kernel.org,
	asas1asas200@gmail.com, regressions@lists.linux.dev
Subject: Re: [REGRESSION] ALSA: hda/realtek: Higole F9B fixup breaks speakers on unrelated Apollo Lake laptop (10ec:111e)
Date: Sun, 22 Nov 2026 09:02:17 +0100	[thread overview]
Message-ID: <87cxrx70qe.wl-tiwai@suse.de> (raw)
In-Reply-To: <CACGjpbf73XdwrG34ffjmYU3ZYuStPpFNXf9w+AGYcTmLyHCN_w@mail.gmail.com>

On Sat, 21 Nov 2026 18:24:47 +0100,
Cristian-Marian Fotache wrote:
> 
> Hello,
> 
> I would like to report an internal speaker regression affecting an
> Intel Apollo Lake laptop with a Realtek ALC269VC codec.
> 
> The regression appears to be related to the following upstream commit:
> 
> 2a8ee6897302ca12ff39e8eb180503956797c60c
> ALSA: hda/realtek: Fix silent speaker on Higole F9B
> 
> Original patch:
> https://patch.msgid.link/20260923141344.1543158-1-asas1asas200@gmail.com
> 
> Affected hardware
> 
> CPU: Intel Celeron J3455 (Apollo Lake)
> 
> PCI audio controller: 8086:5a98
> 
> PCI subsystem ID: 10ec:111e
> 
> Codec: Realtek ALC269VC
> 
> Codec vendor ID: 0x10ec0269
> 
> Codec subsystem ID: 0x10ec111e
> 
> Codec revision: 0x100202
> 
> DMI baseboard product: SCHNEIDER
> 
> BIOS: American Megatrends AN12DV201 (2021-01-25)
> 
> Distribution: Linux Mint Xfce
> 
> This laptop is not a Higole F9B, but shares its subsystem identifier.
> 
> Regression testing
> 
> 6.14.0-37-generic: internal speakers work.
> 
> 7.0.0-43-generic (#43~24.04.1-Ubuntu): internal speakers are silent.
> 
> 7.0.0-43-generic with snd_hda_intel.model=nofixup: internal speakers work again.
> 
> The workaround has also been verified across a reboot.
> 
> Diagnostic findings
> 
> The affected kernel reports:
> 
> ALC269VC: picked fixup  for PCI SSID 10ec:111e
> 
> Comparing the complete codec dumps from the working and affected
> kernels reveals changes to the connection configurations of nodes 0x14
> and 0x15.
> 
> For speaker pin 0x14, the working kernel reports:
> 
> Connection: 2
>   0x0c 0x0d*
> 
> The affected kernel reports:
> 
> Connection: 2
>   0x0c* 0x0d
> In-driver Connection: 1
>   0x0c
> 
> Node 0x15 also receives the restricted in-driver connection list.
> 
> The captured ALSA mixer configurations from the working and affected
> kernels are byte-for-byte identical.
> 
> When booting with snd_hda_intel.model=nofixup, the kernel reports:
> 
> ALC269VC: picked no fixup (nofixup specified)
> 
> The original connection topology is restored, and the internal
> speakers produce sound again.
> 
> Suspected cause
> 
> The subsystem-ID-based Higole F9B quirk appears to match unrelated
> hardware that requires a different audio routing configuration.
> 
> A more restrictive hardware match may be needed to preserve the
> original Higole F9B fix without introducing a regression on other
> devices sharing subsystem ID 10ec:111e.
> 
> I have not performed a full kernel bisection or independently tested
> the suspected commit in an upstream kernel build.
> 
> Comparative codec dumps, mixer configurations, kernel logs, PCI
> information and DMI information are included in the attached
> diagnostic archive. A separate unified codec diff is also attached.
> 
> I would be happy to provide additional diagnostics, test proposed
> patches, or assist in developing a correction that preserves
> functionality on both machines.
> 
> The investigation and preparation of this report were assisted by an
> LLM. All physical hardware testing and diagnostic captures were
> performed directly on the affected machine.
> 
> Thank you for your time.
> 
> Best regards,
> Cristian-Marian Fotache

Thanks for the report.
Unless we come with a quick fix, I'm going to revert the commit.

Yu-Hsiang, could you check whether the quirk can be replaced with
HDA_QUIRK_ENTRY() with the codec ID instead?  This could be more
unique than PCI SSID.


thanks,

Takashi

      reply	other threads:[~2026-11-22  8:02 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-11-21 17:24 [REGRESSION] ALSA: hda/realtek: Higole F9B fixup breaks speakers on unrelated Apollo Lake laptop (10ec:111e) Cristian-Marian Fotache
2026-11-22  8:02 ` Takashi Iwai [this message]

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=87cxrx70qe.wl-tiwai@suse.de \
    --to=tiwai@suse.de \
    --cc=asas1asas200@gmail.com \
    --cc=cristianmarianfotache@gmail.com \
    --cc=linux-sound@vger.kernel.org \
    --cc=perex@perex.cz \
    --cc=regressions@lists.linux.dev \
    --cc=tiwai@suse.com \
    /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