Audio volume abnormal in 6.10.10

Description:

After upgrade to kernel 6.10.10.arch1-1, the audio volume of the speaker suddenly become too loud. At 0.03 (3%), it sounds like 0.5 before the update. And 0.1 is already like 1. But 1 is still 1.

The output of journelctl seems normal. The PipeWire and WirePlumber configuration is never changed. Downgrading to 6.10.9.arch1-2 fix the issue.

Additional info:

• Laptop model: System76 Pangolin (pang14)
• Audio device: Advanced Micro Devices, Inc. [AMD] Family 17h/19h HD Audio Controller
• Tested Broken kernel version: 6.10.10.arch1-1, lts-6.6.51-1, mainline-6.11rc7-1, latest upstream

Does not affect bluetooth devices.

added priority3-normal severity4-low statusunconfirmed labels

assigned to @freswa, @anthraxx, @toolybird, and @gromit

This is the first package from the bisection, please test and report back whether it works or not:

sudo pacman -U https://pkgbuild.com/\~gromit/linux-bisection-kernels/linux-v6.10.9.r187.ge454476-1-x86_64.pkg.tar.zst

The issue persist

Here is the next one:

sudo pacman -U https://pkgbuild.com/\~gromit/linux-bisection-kernels/linux-v6.10.9.r93.g3e0a295-1-x86_64.pkg.tar.zst

The issue persist

Please try:

sudo pacman -U https://pkgbuild.com/\~gromit/linux-bisection-kernels/linux-v6.10.9.r46.gd81ef42-1-x86_64.pkg.tar.zst

The issue persist

Please try:

sudo pacman -U https://pkgbuild.com/\~gromit/linux-bisection-kernels/linux-v6.10.9.r23.g028bd7b-1-x86_64.pkg.tar.zst

The issue persist

added scoperegression label

Please try:

sudo pacman -U https://pkgbuild.com/\~gromit/linux-bisection-kernels/linux-v6.10.9.r11.g54e3f7d-1-x86_64.pkg.tar.zst

The issue persist

Please try:

sudo pacman -U https://pkgbuild.com/\~gromit/linux-bisection-kernels/linux-v6.10.9.r5.gf7fe95f-1-x86_64.pkg.tar.zst

No issue ️

changed the description

Please try:

sudo pacman -U https://pkgbuild.com/\~gromit/linux-bisection-kernels/linux-v6.10.9.r8.gecdbe8a-1-x86_64.pkg.tar.zst

No issue

Please try:

sudo pacman -U https://pkgbuild.com/\~gromit/linux-bisection-kernels/linux-v6.10.9.r10.g85e5f24-1-x86_64.pkg.tar.zst

No issue

Please also test this mainline kernel which has the commit in question reverted:

sudo pacman -U https://pkgbuild.com/\~gromit/linux-bisection-kernels/linux-mainline-6.11-1.1-x86_64.pkg.tar.zst

54e3f7d0a17f5c11a4fa76cc502e12c2fcded623 is the first bad commit
commit 54e3f7d0a17f5c11a4fa76cc502e12c2fcded623
Author: Christoffer Sandberg <cs@tuxedo.de>
Date:   Tue Aug 27 12:25:40 2024 +0200

    ALSA: hda/conexant: Add pincfg quirk to enable top speakers on Sirius devices
    
    commit 4178d78cd7a86510ba68d203f26fc01113c7f126 upstream.
    
    The Sirius notebooks have two sets of speakers 0x17 (sides) and
    0x1d (top center). The side speakers are active by default but
    the top speakers aren't.
    
    This patch provides a pincfg quirk to activate the top speakers.
    
    Signed-off-by: Christoffer Sandberg <cs@tuxedo.de>
    Signed-off-by: Werner Sembach <wse@tuxedocomputers.com>
    Cc: stable@vger.kernel.org
    Link: https://patch.msgid.link/20240827102540.9480-1-wse@tuxedocomputers.com
    Signed-off-by: Takashi Iwai <tiwai@suse.de>
    Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

 sound/pci/hda/patch_conexant.c | 11 +++++++++++
 1 file changed, 11 insertions(+)

Click to see bisect log

git bisect start
# status: waiting for both good and bad commits
# bad: [049be94099ea5ba8338526c5a4f4f404b9dcaf54] Linux 6.10.10
git bisect bad 049be94099ea5ba8338526c5a4f4f404b9dcaf54
# status: waiting for good commit(s), bad commit known
# good: [1611860f184a2c9e74ed593948d43657734a7098] Linux 6.10.9
git bisect good 1611860f184a2c9e74ed593948d43657734a7098
# bad: [e454476c44524024fd98e3fea1a488c8fca45bea] bpf, net: Fix a potential race in do_sock_getsockopt()
git bisect bad e454476c44524024fd98e3fea1a488c8fca45bea
# bad: [3e0a295002822e35c2b45d0a139ceb70d5a5d0a0] wifi: ath12k: fix uninitialize symbol error on ath12k_peer_assoc_h_he()
git bisect bad 3e0a295002822e35c2b45d0a139ceb70d5a5d0a0
# bad: [d81ef42faf969e7d401a914e01dfeafb583f6077] clk: starfive: jh7110-sys: Add notifier for PLL0 clock
git bisect bad d81ef42faf969e7d401a914e01dfeafb583f6077
# bad: [028bd7b41f56d80750c40f77d933eff65ed2af18] irqchip/gic-v2m: Fix refcount leak in gicv2m_of_init()
git bisect bad 028bd7b41f56d80750c40f77d933eff65ed2af18
# bad: [54e3f7d0a17f5c11a4fa76cc502e12c2fcded623] ALSA: hda/conexant: Add pincfg quirk to enable top speakers on Sirius devices
git bisect bad 54e3f7d0a17f5c11a4fa76cc502e12c2fcded623
# good: [f7fe95f40c85311c98913fe6ae2c56adb7f767a7] net: microchip: vcap: Fix use-after-free error in kunit test
git bisect good f7fe95f40c85311c98913fe6ae2c56adb7f767a7
# good: [ecdbe8ac86fb5538ccc623a41f88ec96c7168ab9] KVM: x86: Acquire kvm->srcu when handling KVM_SET_VCPU_EVENTS
git bisect good ecdbe8ac86fb5538ccc623a41f88ec96c7168ab9
# good: [85e5f2451853c14429347fc3f7e46ff7850b5b0b] KVM: SVM: Don't advertise Bus Lock Detect to guest if SVM support is missing
git bisect good 85e5f2451853c14429347fc3f7e46ff7850b5b0b
# first bad commit: [54e3f7d0a17f5c11a4fa76cc502e12c2fcded623] ALSA: hda/conexant: Add pincfg quirk to enable top speakers on Sirius devices

No issues! Everything seems fine!

Would you like to write the bugreport to the upstream kernel devs or should I take care of that? If I should do it it would be good to get a Email adress and name for you so I can credit you with Reported-by: Name <email> in the bugreport. If you write the bugreport it would be nice if you could credit me with Bisected-by: Christian Heusel <christian@heusel.eu> for my bisection efforts

See https://lore.kernel.org/all/0bf3f2f0-0fc6-4ba5-a420-c0874ef82d64@heusel.eu for such an example bugreport

Also sensible information to include would maybe be a full dmesg of a working boot and one where the issue exists, idk what else would be relevant

Hi, I have been thinking about this, and thought I would kindly ask you to write the bugreport. You can credit me as Jerry <jerryluo225@gmail.com> if you want.

Full demsgs of 6.10.10.arch1-1 and 6.11-1.1 with bad commit reverted are here. Please let me know if there is any other information you need.

Thank you again for your help and hard work. You guys are truly amazing!

Could you send the dmesg of 6.11-1 as the bad example instead? The kernel devs usually work against mainline

Sure! Here is the dmesg of 6.11.arch1-1 from core-testing

mentioned in issue #79 (closed)

I have posted the upstream issue: https://lore.kernel.org/all/c930b6a6-64e5-498f-b65a-1cd5e0a1d733@heusel.eu/

@jamney did you see that upstream already saw the report and asked for more information? Please post the output from alsa-info.sh --no-upload (installable via extra/alsa-utils) from a working and non-working boot either here or directly to the mailing list

Thanks for letting me know! I have received the email, will be responding in the mailing list soon.

added statuswaiting-upstream label and removed statusunconfirmed label

assigned to @heftig, @dvzrv, and @tpowa and unassigned @freswa, @toolybird, and @gromit

Please try the following:

sudo pacman -U https://pkgbuild.com/\~gromit/linux-bisection-kernels/linux-mainline-6.12rc1-1.2-x86_64.pkg.tar.zst

It's the latest mainline release with the patch from https://lore.kernel.org/all/87seteli51.wl-tiwai@suse.de applied

Thanks!

Could you check if the issue is still present with 6.12-rc1? Just to make sure that the fix is not unrelated

sudo pacman -U https://pkgbuild.com/\~gromit/linux-bisection-kernels/linux-mainline-6.12rc1-1-x86_64.pkg.tar.zst

Yes, the issue also present in this version

Fixed by b3ebb007060f ("ALSA: hda/conexant: Fix conflicting quirk for System76 Pangolin") which was backported to all stable kernels.

added resolutionupstream statusconfirmed labels and removed statuswaiting-upstream label

closed