No sound on Lenovo Yoga Slim 7 14IPH11 (rt722-sdca, Panther Lake): SoundWire -61 ENODATA on all links

[Gutz78]

On a Lenovo Yoga Slim 7 14IPH11 (Intel , PCI device 8086:e428, subsystem 17aa:3839), audio playback fails to open on all SoundWire-connected endpoints (speaker amp, headset jack, mic). The sound card registers correctly and exposes the expected PCM devices via aplay -l, but attempting to open a stream fails.

System info

Laptop: Lenovo Yoga Slim 7 14IPH11 (also sold as Yoga Slim Ultra 14IPH11)
CPU/Platform: Intel Panther Lake
Audio codec: Realtek rt722-sdca (SoundWire, SDCA), confirmed via dmesg: rt722-sdca sdw:0:3:025d:0722:01
BIOS: tested on both TECN25WW (30 Dec 2025) and TECN28WW (20 May 2026) - see BIOS test result below
OS: Ubuntu 26.04 LTS ("resolute")
Kernel tested: 7.0.0-29-generic (Ubuntu stock) and 7.1.5 (mainline PPA build)

To Reproduce

1. Fresh install of Ubuntu 26.04 on the 14IPH11.
2. Run: speaker-test -D hw:0,2 -c 2 -t wav

Behavior on kernel 7.0.0-29-generic

Sound card registers (aplay -l shows Jack Out, Speaker, HDMI1-3, Deepbuffer Jack Out). Attempting playback fails with:

Unable to set hw params for playback: No data available
Setting of hwparams failed: No data available

dmesg shows repeated:

rt722-sdca sdw:0:3:025d:0722:01: DPN_PortCtrl register write failed for port 1
soundwire sdw-master-0-3: Program transport params failed: -61
soundwire sdw-master-0-3: Program params failed: -61
SDW3-Playback-SimpleJack: ASoC error (-61): at snd_soc_link_prepare() on SDW3-Playback-SimpleJack

Same failure pattern on SDW3-Playback-SmartAmp and SDW3-Capture-SmartMic links.

Also present at boot, before audio errors begin:

ACPI BIOS Error (bug): Failure creating named object [_SB.AUDC], AE_ALREADY_EXISTS
ACPI BIOS Error (bug): Failure creating named object [_SB.BGNV], AE_ALREADY_EXISTS

(multiple other AE_ALREADY_EXISTS objects also logged, including _SB.BGMA, _SB.BGMS, _SB.BGIA, _SB.PC00.PTMA, _SB.PC00.PTMS, _SB.PC00.PTIA, and numerous _SB.PC00.XHCI.RHUB.* USB port objects)

Behavior on kernel 7.1.5 (mainline)

No sound card registers at all (aplay -l returns "no soundcards found"). dmesg shows the topology loader selecting a widget name that doesn't match the actual codec:

sof_sdw sof_sdw: ASoC: Failed to add route rt721 SPK(*) -> Speaker
SDW3-Playback-SmartAmp: ASoC error (-19): at snd_soc_link_init() on SDW3-Playback-SmartAmp
sof_sdw: error -ENODEV: snd_soc_register_card failed -19

Note: hardware reports as rt722, but the topology/route lookup references rt721 - appears to be a naming/topology-selection mismatch specific to this kernel build.

BIOS update test result

Another user with the identical laptop model (Arch Linux forum, https://bbs.archlinux.org/viewtopic.php?id=313255) reported the same -61 failure signature and speculated the root cause was a bad BIOS with duplicate ACPI entities, noting it worked for them on kernel 7.2.0-rc3.

To test this hypothesis directly, I updated the BIOS from TECN25WW to TECN28WW (Lenovo's latest at time of testing, released 20 May 2026). Result: no change. On kernel 7.0.0-29-generic, post-update, the identical ACPI error is still present:

ACPI BIOS Error (bug): Failure creating named object [_SB.AUDC], AE_ALREADY_EXISTS

And speaker-test fails identically to before the update:

Unable to set hw params for playback: No data available
Setting of hwparams failed: No data available

The TECN28WW changelog, and the two intervening versions (TECN26WW, TECN27WW), list no audio or ACPI-related fixes - only unrelated changes (NHLT DMic location update, EDID override for external displays, memory SPD data, SAR table updates). So in this case the BIOS does not appear to be the fix, despite the other user's report - this looks like it needs a kernel/driver-side fix after all, though the ACPI duplicate-object bug may still be a genuine (separate, cosmetic-only-so-far) BIOS defect worth noting.

ALSA UCM configuration check

For completeness: alsa-ucm-conf (1.2.15.3-1ubuntu1.5) is installed and does include a profile for this exact codec:

/usr/share/alsa/ucm2/codecs/rt722
/usr/share/alsa/ucm2/sof-soundwire/rt722.conf

So the userspace UCM layer is correctly provisioned for this hardware; the failure occurs earlier, at the kernel/SOF stream-open level, before UCM would come into play.

Expected behavior

Audio playback/capture works via the onboard speakers, headset jack, and microphone array.

Additional context

Happy to provide further dmesg output, sof-logger traces, or test proposed patches - this is a daily-driver machine so I can iterate on this over time. Planning to test kernel 7.2 once it's out of release-candidate status, per the other user's report, and will follow up with results.

[Gutz78]

Further update: also tested kernel 7.2-rc5 (mainline PPA build) specifically because the other user on the Arch thread linked above reported success on 7.2.0-rc3. Unfortunately, on this exact machine, 7.2-rc5 shows the identical failure to 7.0.0-29-generic:

rt722-sdca sdw:0:3:025d:0722:01: DPN_PortCtrl register write failed for port 1
SDW3-Playback-SimpleJack: ASoC error (-61): at snd_soc_link_prepare() on SDW3-Playback-SimpleJack

speaker-test still fails identically:

Unable to set hw params for playback: No data available
Setting of hwparams failed: No data available

So across three kernel versions now (7.0.0-29-generic, 7.1.5, 7.2-rc5) and two BIOS versions (TECN25WW, TECN28WW), the issue persists in some form on every combination tested. Summary of results across all testing so far:

7.0.0-29-generic, BIOS TECN25WW: card registers, playback fails -61 ENODATA
7.0.0-29-generic, BIOS TECN28WW: identical to above, no change from BIOS update
7.1.5, BIOS TECN28WW: card fails to register at all, topology references rt721 instead of rt722
7.2-rc5, BIOS TECN28WW: card registers, playback fails -61 ENODATA (same as 7.0.0-29)

Given the other user's report of success on 7.2.0-rc3 specifically, it's possible whatever fixed it for them was reverted or altered between rc3 and rc5, or their success depended on a factor not present here (different board variant, or a different combination of fixes). Happy to test 7.2.0-rc3 specifically if that would help narrow this down, or any other specific build/patch a maintainer would find useful. This is a daily-driver machine so I can keep testing over time.

sof-bug-report-final-followup.txt