Repository navigation
Description
System
- Laptop: HP 250 15.6 inch G9 Notebook PC, board 89A0, SKU 6S7Q2EA#ABF
- BIOS: F.72 (latest available from HP)
- Audio controller: Intel Alder Lake PCH-P HDA [8086:51c8], subsystem 103c:89a0
- Codecs: Realtek ALC236 (addr 0, SSID 103c:89a0) + Intel Alderlake-P HDMI (addr 2); 2 DMICs in NHLT
- Distro: Fedora 45 Workstation
- Kernels tested: 7.3.3-300.fc45, 7.2.11-300.fc45
- Firmware: alsa-sof-firmware 2026.09.2, linux-firmware 20261019, SOF fw 2:2:0-57864, topology sof-hda-generic-2ch.tplg
- Userspace: PipeWire 1.6.11, WirePlumber 0.5.19, alsa-ucm-conf 1.2.16.1
Note: ALC236: picked fixup for PCI SSID 103c:89a0 — no quirk entry exists for this SSID.
Problem 1: 3.5 mm headphone / headset jack is not detected
With a headset plugged in (headset works on other machines), Headphone Jack and Mic Jack controls stay off, the UCM Headphones port is not available, and the desktop never offers the headset. Forcing the Headphones UCM profile makes it appear only until WirePlumber re-evaluates availability. Detection worked briefly once (Headphone Jack = on for about 1 minute) and then stopped again.
Key observation: the ALC236 answers every verb with the same value
Shortly after boot (within about 1 minute of a cold boot), the Realtek codec returns the constant 0x0000fe05 to every verb, on any node:
$ sudo hda-verb /dev/snd/hwC0D0 0x21 GET_PIN_SENSE 0
nid = 0x21, verb = 0xf09, param = 0x0
value = 0xfe05
The same value comes back for GET_CONFIG_DEFAULT, GET_PIN_WIDGET_CONTROL, GET_UNSOLICITED_RESPONSE and GET_POWER_STATE on nodes 0x21 and 0x19. /proc/asound/cardN/codec#0 decodes that constant as AFG parameters and lists no widget nodes:
Default PCM:
rates [0xe05]: 8000 16000 176400 192000 384000
formats [0x5]: PCM AC3
Default Amp-In caps: ofs=0x05, nsteps=0x7e, stepsize=0x00, mute=0
...
Invalid AFG subtree
The HDMI codec (addr 2) on the same link answers normally (13 nodes listed). During probe, the ALC236 answers correctly: autoconfig finds line_outs=1 (0x14) type:speaker, hp_outs=1 (0x21), Mic=0x19, and the internal speaker keeps working afterwards. Pin sense always lacks the presence bit, so jack detection reads "unplugged".
What does not change the behaviour
| Variation | Result |
|---|---|
| Cold boot (power off 30 s) vs. warm reboot | Same |
SOF driver vs. legacy snd_hda_intel (snd_intel_dspcfg.dsp_driver=1) |
Same (legacy: hdaudioC1D0, same 0xfe05 replies, jack off) |
| Kernel 7.3.3 vs. 7.2.11 | Same |
Codec runtime PM disabled (power/control=on on ehdaudio0D0) |
Same; codec was already active |
| HP UEFI hardware diagnostics, audio test | Passes |
Problem 2: SOF probe hangs after a warm reboot (cold boot is fine)
After reboot, no sound card is created. The probe stops right after use msi interrupt mode and never prints hda codecs found:
snd_soc_avs 0000:00:1f.3: Digital mics found on Skylake+ platform, using SOF driver
sof-audio-pci-intel-tgl 0000:00:1f.3: enabling device (0000 -> 0002)
sof-audio-pci-intel-tgl 0000:00:1f.3: DSP detected with PCI class/subclass/prog-if 0x040100
sof-audio-pci-intel-tgl 0000:00:1f.3: bound 0000:00:02.0 (ops intel_audio_component_bind_ops [i915])
sof-audio-pci-intel-tgl 0000:00:1f.3: use msi interrupt mode
<nothing further>
On a good boot the next lines are hda codecs found, mask 5 and using HDA machine driver skl_hda_dsp_generic now. The probe work never completes: writing to the device's sysfs remove blocks forever in D state:
[<0>] __flush_work+0x315/0x4c0
[<0>] cancel_work_sync+0x5e/0x80
[<0>] snd_sof_device_remove+0x29/0x170 [snd_sof]
[<0>] sof_pci_remove+0x21/0x4f [snd_sof_pci]
[<0>] pci_device_remove+0x4a/0xb0
[<0>] device_release_driver_internal+0x191/0x240
[<0>] pci_stop_bus_device+0x80/0xa0
[<0>] pci_stop_and_remove_bus_device_locked+0x1a/0x30
[<0>] remove_store+0x83/0xa0
A kworker/u49:3+i915_flip was also in D state for the whole boot. Shutdown then hangs with rcu_tasks grace period number 9 (since boot) is 10184 jiffies old.
Workaround that avoids the hang: unbind 0000:00:1f.3 from its driver in a systemd unit ExecStop before shutdown or reboot. With that in place, the warm reboots tested so far (on both kernels) brought the card up. Problem 1 remains either way.
Possibly related: #5922 (HP 15-dy2xxx, ALC236, SSID 103c:87fe, same jack-detection symptom).
Hypothesis
Both problems point to the ALC236 on this board ending up in a state where it stops responding properly to verbs: after a warm reset (codec not power-cycled), and also at runtime shortly after boot. This may need a board-specific quirk or fixup for 103c:89a0 (similar HP 250/255 G8–G10 models have entries). I'm happy to test patches or a hda_model= override.
Kernel log excerpts
Bad warm-reboot boot vs. good boot vs. legacy driver
=== BAD warm-reboot boot (kernel 7.3.3): probe stops after 'use msi interrupt mode', no card created
[4.468028] snd_soc_avs 0000:00:1f.3: Digital mics found on Skylake+ platform, using SOF driver
[4.605041] sof-audio-pci-intel-tgl 0000:00:1f.3: enabling device (0000 -> 0002)
[4.605363] sof-audio-pci-intel-tgl 0000:00:1f.3: DSP detected with PCI class/subclass/prog-if 0x040100
[4.605529] sof-audio-pci-intel-tgl 0000:00:1f.3: bound 0000:00:02.0 (ops intel_audio_component_bind_ops [i915])
[4.612042] sof-audio-pci-intel-tgl 0000:00:1f.3: use msi interrupt mode
=== GOOD boot (kernel 7.3.3)
[4.568019] snd_soc_avs 0000:00:1f.3: Digital mics found on Skylake+ platform, using SOF driver
[4.706012] sof-audio-pci-intel-tgl 0000:00:1f.3: enabling device (0000 -> 0002)
[4.706423] sof-audio-pci-intel-tgl 0000:00:1f.3: DSP detected with PCI class/subclass/prog-if 0x040100
[4.706637] sof-audio-pci-intel-tgl 0000:00:1f.3: bound 0000:00:02.0 (ops intel_audio_component_bind_ops [i915])
[4.713023] sof-audio-pci-intel-tgl 0000:00:1f.3: use msi interrupt mode
[4.761098] sof-audio-pci-intel-tgl 0000:00:1f.3: hda codecs found, mask 5
[4.761309] sof-audio-pci-intel-tgl 0000:00:1f.3: using HDA machine driver skl_hda_dsp_generic now
[4.761408] sof-audio-pci-intel-tgl 0000:00:1f.3: NHLT device BT(0) detected, ssp_mask 0x4
[4.761489] sof-audio-pci-intel-tgl 0000:00:1f.3: BT link detected in NHLT tables: 0x4
[4.761579] sof-audio-pci-intel-tgl 0000:00:1f.3: DMICs detected in NHLT tables: 2
[4.789630] sof-audio-pci-intel-tgl 0000:00:1f.3: Firmware file: intel/sof/sof-adl.ri
[4.789863] sof-audio-pci-intel-tgl 0000:00:1f.3: Topology file: intel/sof-tplg/sof-hda-generic-2ch.tplg
[4.805070] sof-audio-pci-intel-tgl 0000:00:1f.3: Firmware info: version 2:2:0-57864
[4.805260] sof-audio-pci-intel-tgl 0000:00:1f.3: Firmware: ABI 3:22:1 Kernel ABI 3:23:1
[4.924004] sof-audio-pci-intel-tgl 0000:00:1f.3: Topology: ABI 3:22:1 Kernel ABI 3:23:1
[4.944131] snd_hda_codec_alc269 ehdaudio0D0: ALC236: picked fixup for PCI SSID 103c:89a0
[4.956020] snd_hda_codec_alc269 ehdaudio0D0: autoconfig for ALC236: line_outs=1 (0x14/0x0/0x0/0x0/0x0) type:speaker
[4.956417] snd_hda_codec_alc269 ehdaudio0D0: speaker_outs=0 (0x0/0x0/0x0/0x0/0x0)
[4.956567] snd_hda_codec_alc269 ehdaudio0D0: hp_outs=1 (0x21/0x0/0x0/0x0/0x0)
[4.956693] snd_hda_codec_alc269 ehdaudio0D0: mono: mono_out=0x0
[4.956801] snd_hda_codec_alc269 ehdaudio0D0: inputs:
[4.956904] snd_hda_codec_alc269 ehdaudio0D0: Mic=0x19
[5.001044] skl_hda_dsp_generic skl_hda_dsp_generic: hda_dsp_hdmi_build_controls: no PCM in topology for HDMI converter 3
[5.023325] input: sof-hda-dsp Headphone as /devices/pci0000:00/0000:00:1f.3/skl_hda_dsp_generic/sound/card0/input16
=== Legacy driver boot (snd_intel_dspcfg.dsp_driver=1)
[4.821014] snd_hda_intel 0000:00:1f.3: enabling device (0000 -> 0002)
[4.821263] snd_hda_intel 0000:00:1f.3: bound 0000:00:02.0 (ops intel_audio_component_bind_ops [i915])
[4.868013] snd_hda_codec_alc269 hdaudioC1D0: ALC236: picked fixup for PCI SSID 103c:89a0
[4.868278] snd_hda_codec_alc269 hdaudioC1D0: autoconfig for ALC236: line_outs=1 (0x14/0x0/0x0/0x0/0x0) type:speaker
[4.868450] snd_hda_codec_alc269 hdaudioC1D0: hp_outs=1 (0x21/0x0/0x0/0x0/0x0)
[4.868675] snd_hda_codec_alc269 hdaudioC1D0: Mic=0x19
Attachments
- alsa-info.txt: output of
alsa-info.sh --no-upload(kernel 7.2.11, SOF driver)
Activity
LifeNum-MFE commentedon Nov 12, 2026
Correction: the blank in
picked fixup for PCI SSID 103c:89a0is because CONFIG_SND_DEBUG_VERBOSE is off — the existing0x89a0 "HP Laptop 15-dw4xxx"entry (ALC236_FIXUP_HP_MUTE_LED_COEFBIT2) is applied, so the "no quirk entry exists" line in my report is wrong.New test: booting with
snd_sof_intel_hda_generic.hda_model=nofixup(dmesg:ALC236: picked no fixup (nofixup specified)) changes nothing — the codec still returns 0xfe05 to all verbs, and Headphone/Mic Jack stayoffwith a headset plugged in.