ASoC: Intel: soc-acpi-intel-ptl-match: Remove unnecessary cs42l43 match

[charleskeepax]

For PTL onwards we are intending to rely on function topologies, rather than using a match table for each system type. Remove this unnecessary match table entry. Having the match entries can mean that systems match when they should use function topologies instead, resulting in incorrect audio configurations. Although, admittedly this is not too likely with this 6x amp configuration as those are quite rare, but best to follow best practice.

Note there is a single remaining match which does currently need a match as it contains custom magic in the topology. More work will be required to figure out how we guard off situations like that from interfering with more generic systems.

[Copilot]

Pull request overview

This PR updates Intel PTL SoundWire machine matching to prefer function topology selection over a per-platform match entry, and adds a SoundWire DMI quirk to ignore a ghost RT722 ACPI device on specific PTL systems to avoid incorrect enumeration.

Changes:

• Remove the PTL SoundWire machine match entry for the cs42l43 + 6x cs35l56 configuration to avoid preempting function-topology based selection.
• Add a new DMI-based _ADR remap quirk to drop a ghost RT722 device (by remapping its _ADR to 0) on select ASUS/Lenovo PTL platforms.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 2 comments.

File	Description
sound/soc/intel/common/soc-acpi-intel-ptl-match.c	Removes an explicit cs42l43+cs35l56x6 match entry so PTL systems rely on function topologies instead.
drivers/soundwire/dmi-quirks.c	Adds a DMI quirk table entry to ignore a ghost Realtek RT722 ACPI device on specific PTL systems.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[lgirdwood]

LGTM, but needs @bardliao

[Pegorim]

Tested on ASUS Zenbook S14 UX5406AA (Panther Lake, Core Ultra 9 386H), BIOS v303, against torvalds linux v7.1-rc2 with this PR's current HEAD (a5bec626, the 3-commit version with the shared ghost_realtek quirk table and the Chrome-only machine_check) plus PR #5728. Audio works end-to-end — confirming the DMI-quirk approach as a clean replacement for PR #5754.

Boot evidence (journalctl -k -b 0 | grep -iE "sof|sdw|cs35l56|cs42l43|rt722"):

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 1 mfg_id 0x01fa part_id 0x3556 version 0x3
sof-audio-pci-intel-ptl 0000:00:1f.3:   link 1 mfg_id 0x01fa part_id 0x3556 version 0x3
sof-audio-pci-intel-ptl 0000:00:1f.3:   link 2 mfg_id 0x01fa part_id 0x3556 version 0x3
sof-audio-pci-intel-ptl 0000:00:1f.3:   link 2 mfg_id 0x01fa part_id 0x3556 version 0x3
sof-audio-pci-intel-ptl 0000:00:1f.3:   link 3 mfg_id 0x01fa part_id 0x4243 version 0x3
sof-audio-pci-intel-ptl 0000:00:1f.3: Use SoundWire default machine driver with function topologies
...
sof-audio-pci-intel-ptl 0000:00:1f.3: Using function topologies instead intel/sof-ipc4-tplg/sof-ptl-dummy.tplg
sof-audio-pci-intel-ptl 0000:00:1f.3: loading topology 0: intel/sof-ipc4-tplg/sof-sdca-2amp-id2.tplg
sof-audio-pci-intel-ptl 0000:00:1f.3: loading topology 1: intel/sof-ipc4-tplg/sof-sdca-jack-id0.tplg
sof-audio-pci-intel-ptl 0000:00:1f.3: loading topology 2: intel/sof-ipc4-tplg/sof-sdca-mic-id4.tplg
sof-audio-pci-intel-ptl 0000:00:1f.3: loading topology 3: intel/sof-ipc4-tplg/sof-hdmi-pcm5-id5.tplg
cs35l56 sdw:0:1:01fa:3556:01:2: DSP1: cirrus/cs35l56-b0-dsp1-misc-10431464-l1u2.bin (v1): v3.11.28
cs35l56 sdw:0:2:01fa:3556:01:0: DSP1: cirrus/cs35l56-b0-dsp1-misc-10431464-l2u0.bin (v1): v3.11.28
cs35l56 sdw:0:1:01fa:3556:01:3: DSP1: cirrus/cs35l56-b0-dsp1-misc-10431464-l1u3.bin (v1): v3.11.28
cs35l56 sdw:0:2:01fa:3556:01:1: DSP1: cirrus/cs35l56-b0-dsp1-misc-10431464-l2u1.bin (v1): v3.11.28

What's confirmed:

• DMI match fires: the new ghost_realtek remap entry for ASUS / UX5406AA zeroes the phantom 0x000330025d072201 _ADR. No rt722 traces in dmesg, and the enumerated devices are just the four cs35l56 on links 1/2 plus the cs42l43 on link 3 (i.e. real hardware only).
• All 4 amps load tuning blobs (l1u2, l1u3, l2u0, l2u1) — same coverage as under ASoC: soc-acpi-intel-ptl-match: add ptl_cs42l43_agg_l3_cs35l56_l12_ghost_rt722 #5754, no regression.
• Userspace sees the expected device set: Jack Out/In, Speaker, Microphone, HDMI 1/2/3 — both via aplay -l and Pipewire sinks (HiFi__Headphones__sink, HiFi__Speaker__sink, three HDMI*__sinks).

One observation worth flagging on the path the kernel takes: after the rt722 is removed, the ptl_cs42l43_agg_l3_cs35l56_l2 entry is now skipped on non-Google DMI thanks to the new .machine_check = snd_soc_acpi_intel_no_function_topology, so the kernel falls all the way through to the SoundWire default machine driver with function topologies path. That default path then enumerates everything via function topologies — and that's what makes the four-amp setup work without a per-platform machine entry. So the elegance of this version of #5773 vs. #5754 is that the same DMI quirk, combined with the Chrome-only fence around the static-tplg entry, lets that default path do the heavy lifting on any board sharing the ghost-rt722 pattern.

For my @Pegorim/omarchy-pkgs#90, I'm dropping the prior backport of #5754 and shipping just #5728 + this PR as the linux-ptl audio set, since the result on hardware is identical and the patch surface is much smaller.

cc @itisyb for the Lenovo 83QK/83SF coverage now in the same shared ghost_realtek table.

[charleskeepax]

@Pegorim Thanks very much for the testing there is one slight oddity reported on the Yoga Pro 9 that I want to attempt to track down, then I shall get the series upstreamed.