#16123
Issues/omacom/omarchy/#16123

Oct 19 update left HP EliteBook X Flip G2i with silent audio: TAS2783 UCM profile missing

Openrobbyproc87opened 1d ago

Oct 19 update left HP EliteBook X Flip G2i with silent audio: TAS2783 UCM profile missing

Impact

The October 19, 2026 Omarchy/system update left my HP EliteBook X Flip G2i without audible laptop speaker or wired-headphone output. I discovered this while trying to join a live Teams meeting in Chrome. Troubleshooting consumed meeting time, and Teams showed no usable speaker or microphone in its device picker. A routine update should not leave a previously working laptop without a reliable audio path for a call.

This report documents a reproduced ALSA UCM failure. I have not isolated which individual October 19 package first made the physical outputs silent. I also have not proven that the remaining Teams browser device problem has the same cause.

Machine and update

  • HP EliteBook X Flip G2i 14 inch, Intel Panther Lake, SOF SoundWire card with spk:tas2783, hs:rt712, mic:rt712, and four TAS2783 speaker amps.
  • Current packages: Omarchy 4.0.6-1, linux-omarchy 7.2.8-9, sof-firmware 2026.09.2-1, alsa-ucm-conf 1.2.16.1-1, PipeWire 1:1.6.9-1, WirePlumber 0.5.18-1.
  • October 19 transaction upgraded linux-omarchy 7.2.5-3 → 7.2.8-9, sof-firmware 2025.12.2-1 → 2026.09.2-1, PipeWire 1.6.8 → 1.6.9, and WirePlumber 0.5.17 → 0.5.18. alsa-ucm-conf remained 1.2.16.1-1.
  • The prior known-good boot used SOF firmware 2.14.1.1; the affected boot loaded 2.15.0.1. That timeline does not prove a firmware defect.

Reproduction and evidence

  1. With the stock installed UCM directory, ALSA_CONFIG_UCM2=/usr/share/alsa/ucm2 alsaucm -c hw:0 dump text exits 1:

    could not open configuration file /usr/share/alsa/ucm2/sof-soundwire/tas2783.conf
failed to import hw:0 use case configuration -2
alsaucm: error failed to open sound card hw:0: No such file or directory

  2. /usr/share/alsa/ucm2/sof-soundwire/tas2783.conf is absent from installed alsa-ucm-conf 1.2.16.1-1. The card requires it for its TAS2783 speaker component.

  3. PipeWire initially showed laptop speakers and laptop mic as unmuted defaults. A generated tone reached the PipeWire sink monitor, yet PipeWire and direct ALSA playback tests were inaudible from both laptop speakers and wired headphones. A short laptop mic capture succeeded, so the mic hardware was capturing.

  4. I copied the stock UCM tree to a user-level overlay and added ALSA upstream's sof-soundwire/tas2783.conf and codecs/tas2783/init.conf. Pointing ALSA_CONFIG_UCM2 to that overlay makes the same alsaucm -c hw:0 dump text exit 0 and expose HiFi Speaker, Headphones, and Mic routes.

  5. After WirePlumber loaded the overlay and I selected HiFi, an analog-headphone test tone became audible and a second short laptop-mic capture succeeded. Laptop speakers were not retested after this workaround. Teams browser audio has not been verified as restored.

This differs from #14172, the TAS2783 resume delay, and #12720, the WirePlumber 0.5.17 IPU7 deadlock. This system has WirePlumber 0.5.18 and had audio nodes before the UCM workaround.

Requested fix

Please ship a tested TAS2783 UCM profile through Omarchy's package/update path, or coordinate an alsa-ucm-conf package compatible with the kernel and SOF firmware versions Omarchy distributes. Before shipping that combination to stable, test UCM loading, speaker and analog-headphone playback, built-in mic capture, and browser call device enumeration on this hardware class. A failed UCM import should produce an actionable warning instead of leaving users to discover silent audio in a meeting.

I can supply redacted diagnostics or test a proposed package on this laptop.

Activity

1
robbyproc871d ago

Additional evidence from the earlier October 20 troubleshooting, before the UCM-overlay workaround described above:

  • The system already had a user WirePlumber workaround forcing device.profile = "pro-audio" for the SOF SoundWire card. It assigned friendly names to pro-output-2 (Laptop Speakers), pro-output-0 (Headphone Jack), pro-input-4 (Laptop Microphone), and pro-input-1 (Headset Microphone). Thus the presence of named audio devices did not establish that stock UCM routing was working.
  • Chrome playback streams were actively linked to pro-output-2, and its recording stream was actively linked to pro-input-4. The default speaker sink and Chrome playback streams were unmuted at 100%; the laptop mic was also unmuted. All four ALSA speaker switches (Left Spk, Left Spk2, Right Spk, Right Spk2) were already on.
  • A roughly two-second, 16 kHz mono sample measured normalized RMS 0.026297, peak 0.199860 from the laptop mic, and RMS 0.070047, peak 0.558441 from the speaker sink monitor. These measurements show capture and software playback activity; they do not prove audible physical output or successful end-to-end call audio.

The following recovery attempts did not restore audible laptop-speaker output, with the user confirming silence after each stage:

  1. Suspend/resume the speaker sink using pactl suspend-sink <speaker-sink> 1, then 0.
  2. Restart pipewire, pipewire-pulse, and wireplumber with systemctl --user restart; reselect the laptop speaker default, explicitly unmute it, and set its volume to 80%.
  3. Test generated tones through both pro-output-2 (Speaker) and pro-output-35 (Deepbuffer Speaker). Both playback commands returned exit status 0, but neither was audible.
  4. Cycle all four ALSA speaker switches off and back on, then replay the tone. Output remained silent.

This establishes that the initial failure also affected a local test tone independently of Teams. A successful playback command, active PipeWire stream, or unmuted device was insufficient to verify recovery. It does not identify which updated package caused the failure, and these earlier failed tests do not establish the state of the laptop speakers after the later UCM workaround.