## Summary

On a Dell XPS 13 DX13260 (PCI subsystem ID `1028:0e53`, Panther Lake/Wildcat Lake platform), audio is broken in two distinct, stacked ways:

1. **~50% of boots**: the `sof_sdw` ASoC card fails to instantiate at all due to a race between the two CS35L56 SPI sidecar speaker amps (left/right) probing concurrently. PipeWire falls back to `auto_null` (Dummy Output). `/proc/asound/cards` shows zero cards.
2. **On the boots where the card does instantiate successfully**: both amps report `FIRMWARE_MISSING` / "Calibration disabled due to missing firmware controls" / "Can't read tuning IDs" even though the correct tuning files exist on disk (`cs35l56-b2-dsp1-misc-10280e53-spkid{1,2,3}*` in `linux-firmware-cirrus`). Result: sinks/controls all report healthy (unmuted, full volume, PCM streams active with no xruns), but there is no audible output.

## Hardware

- Model: Dell XPS 13 DX13260
- PCI subsystem ID: `1028:0e53`
- Codec: Cirrus CS42L43 (SoundWire) + 2x Cirrus CS35L56 (SPI sidecar amps, Rev B2 OTP1, fw 4.2.1)
- Kernel: `linux 7.2.3.arch1-3` (stock Arch/Omarchy kernel, not `linux-ptl`)
- `linux-firmware-cirrus 20260810-2`
- Omarchy `4.0.3-1`

## The `dell-xps13-sidecar-amps` package appears to be a no-op on this kernel

I installed `omarchy/dell-xps13-sidecar-amps` (which sets `options snd_soc_sof_sdw quirk=65536` via modprobe.d) hoping it would fix the card-instantiation failures. Kernel log shows:

```
sof_sdw sof_sdw: Overriding quirk 0x10000 => 0x10000
```

i.e. the "override" changes nothing — the kernel already defaults to `0x10000` (`SOC_SDW_SIDECAR_AMPS`) for this SSID natively (consistent with the package's own README noting the upstream fix, commit `efd80de2de9d`, landed in Linux 7.2-rc5, and this system is on 7.2.3). A machine with the override genuinely taking effect would show something like `Overriding quirk 0x1 => 0x10000` instead.

So the quirk package doesn't explain the intermittent success/failure observed here — that appears to be a separate, still-present probe race in the SPI amp bring-up path.

## Reproduction log across consecutive boots (same session, same hardware, no config changes between boots 1-4)

**Boot 1** — card fails to instantiate:
```
cs35l56 spi-cs35l56-right: DSP1: Firmware: 1a00d6 vendor: 0x2 v4.2.1, 42 algorithms
cs35l56 spi-cs35l56-left: Failed to schedule write: -16
cs35l56 spi-cs35l56-left: ASoC error (-16): at snd_soc_pcm_dai_probe() on cs35l56-asp1
sof_sdw sof_sdw: ASoC: failed to instantiate card -16
```
(Right amp DSP loads fine, then left amp fails to schedule a write with `-EBUSY`, and card instantiation aborts entirely.)

**Boot 2** (after installing `dell-xps13-sidecar-amps`) — same failure, opposite amp fails this time:
```
cs35l56 spi-cs35l56-right: Failed to schedule write: -16
cs35l56 spi-cs35l56-right: ASoC error (-16): at snd_soc_pcm_dai_probe() on cs35l56-asp1
sof_sdw sof_sdw: ASoC: failed to instantiate card -16
```

**Boot 3** — card instantiates successfully (`sof-soundwire`, machine string `DellInc.-XPS13DX13260-1.3.0-0865D8`), full PipeWire sink/source set appears, `paplay`/`speaker-test` run with zero errors and sink transitions `SUSPENDED` → `RUNNING`, but:
```
cs35l56 spi-cs35l56-right: FIRMWARE_MISSING
cs35l56 spi-cs35l56-right: Calibration disabled due to missing firmware controls
cs35l56 spi-cs35l56-right: Can't read tuning IDs
cs35l56 spi-cs35l56-left: FIRMWARE_MISSING
cs35l56 spi-cs35l56-left: Calibration disabled due to missing firmware controls
cs35l56 spi-cs35l56-left: Can't read tuning IDs
```
No audible output despite active, unmuted, full-volume playback streams. `AMPL/AMPR CAL_SET_STATUS` mixer controls both read `Unknown` (never progress to `Default`/`Set`).

**Boot 4** — back to the boot-1/2 failure mode (`-16` EBUSY, no card at all, dummy sink).

## Why this doesn't look like a missing-firmware problem

The exact tuning files this chip needs are present on disk:
```
/lib/firmware/cirrus/cs35l56-b2-dsp1-misc-10280e53-spkid1-ampl.bin.zst
/lib/firmware/cirrus/cs35l56-b2-dsp1-misc-10280e53-spkid1-ampr.bin.zst
/lib/firmware/cirrus/cs35l56-b2-dsp1-misc-10280e53-spkid1.wmfw.zst
... (spkid2, spkid3 variants also present)
```
so `linux-firmware-cirrus` already ships the right calibration data for this SSID — the driver just never successfully reads/applies it, which points at a timing/race issue in the SPI amp bring-up (plausibly the same class of bug as the `-16 EBUSY` card-instantiation failure) rather than a packaging gap.

## Ask

Given this is very new hardware (Panther Lake / Wildcat Lake), this may need a kernel-side fix (SPI CS/timing sequencing between the two `cs35l56` sidecar amps) rather than anything fixable from Omarchy's packaging side. Filing here in case others hit the same laptop, and in case `linux-ptl` already has relevant patches that should be considered for the stock `linux` package, or in case this warrants tracking similarly to #5557.

Happy to provide more diagnostics (full `dmesg`, `amixer -c0 contents`, etc.) on request.