[BUG] MECHREVO YAOSHI (Panther Lake): speakers silent — SoundWire codec 0x6245 not supported

[jfdnet]

No sound from internal speakers on MECHREVO YAOSHI (Panther Lake, Core Ultra 7 356H). The card enumerates but only exposes HDMI outputs; PipeWire/WirePlumber ends up with a Dummy Output sink because no analog output device exists.

Root cause

The SOF driver does not recognize the ACPI-reported SoundWire codec and falls back to the HDMI-only HDA machine driver:

sof-audio-pci-intel-ptl 0000:00:1f.3: No SoundWire machine driver found for the ACPI-reported configuration:
sof-audio-pci-intel-ptl 0000:00:1f.3: link 3 mfg_id 0x0496 part_id 0x6245 version 0x3
sof-audio-pci-intel-ptl 0000:00:1f.3: part id 0x6245 is not supported
sof-audio-pci-intel-ptl 0000:00:1f.3: hda codecs found, mask 4
sof-audio-pci-intel-ptl 0000:00:1f.3: using HDA machine driver skl_hda_dsp_generic now

The internal speakers sit on a SoundWire codec on link 3 with mfg_id 0x0496 (manufacturer unknown) and part_id 0x6245, which is absent from the SOF driver's codec table. Only HDMI PCMs are created (/proc/asound/pcm shows 7 HDMI devices, no analog/headphone), so PipeWire falls back to auto_null.

Environment

• DMI: MECHREVO YAOSHI Series / Standard, SKU 0001, BIOS N.1.05MRO13 06/18/2026
• Audio PCI: 8086:e328 Intel Core Ultra (Series 3) HD Audio, subsystem 1d05:30b3 (AIstone Global Limited)
• Kernel: 7.1.8-arch1-3 (Arch Linux)
• sof-firmware: 2025.12.2-1 (ADSPFW 2.14.1.1 boots successfully)
• linux-firmware: 20260810-1
• pipewire 1.6.8, wireplumber 0.5.15

Request

Add the SoundWire codec on link 3 (mfg_id 0x0496 / part_id 0x6245) to the SOF driver codec table, and provide the matching SoundWire machine driver / topology. The DSP itself boots fine (firmware 2.14.1.1 loads, no firmware errors), so this appears to be only a codec/machine matching gap.

Full diagnostics (dmesg + alsa-info): https://gist.github.com/jfdnet/f59e0b9e3263c3f06435eaa622fc06cf

[charleskeepax]

mfg_id 0x0496 / part_id 0x6245, Cirrus Logic

This is not the manufacturer ID for Cirrus, nor is that a valid Cirrus part number. No idea who's part this is but it ain't us.

[jfdnet]

Thanks for the correction — you're right, I misidentified the manufacturer. Cirrus's MIPI MID is 0x01FA; this machine's ACPI reports mfg_id 0x0496 / part_id 0x6245, a different vendor I haven't been able to identify. I've removed the incorrect "Cirrus" references. Any idea which vendor owns MID 0x0496?