Dell XPS 13 (2026, DX13260) on Omarchy

Verdict

Bronze on Omarchy 4.0.4. The 2026 XPS 13, DMI product name XPS 13 DX13260, is a good machine with one bad subsystem and two open annoyances. Internal speakers are the bad subsystem. Across seven issues filed between August and September 2026, the Cirrus CS35L56 amplifiers on this chassis either never load their firmware or the sound card fails to come up at all. A Cirrus driver patch fixes the firmware half in testing, but it is not in the stock Arch kernel or in linux-omarchy 7.2.5-3, so nothing Omarchy ships fixes it today.

Everything else is livable. The 2560x1600 panel gets a text size tuned for it, and the machine installs and runs. If you want a laptop where audio is handled, the XPS 14 and 16 are the models Omarchy actually ships speaker tunings for.

What works

Display text size is pre-adjusted. install/user/hardware/dell/xps13-text-scaling.sh matches DMI DX13260 and calls omarchy-display-text-size 11, because the panel renders the default a touch large.

Suspend works on the stock Arch kernel. In issue #12190, thirteen of thirteen suspends completed on 7.2.3-arch1-3 on this exact model.

The Synaptics haptic touchpad package is an XPS-wide enablement: omarchy-hw-dell-xps-haptic-touchpad matches any XPS with the i2c-VEN_06CB:00 device, and click strength is then adjustable from the menu under Trigger, Hardware, Touchpad Haptics. See the keyboard, mouse and trackpad chapter of the manual. No DX13260 report mentions the touchpad either way, so it is listed as unknown rather than confirmed.

Audio paths that avoid the amps are fine where they were tested. A USB-C headset works in #11320, headphones through the CS42L43 codec are unaffected in #9687, and the internal mics behaved normally in #7427. Note that #11320 says the DX13260 has two USB-C ports and no analog jack, so headphones means USB-C or a dock.

What breaks

Speakers. There are two SKUs and both are broken in different ways. On SKU 0E53 (audio subsystem 1028:0e53) the amps report their speaker ID as 0 and request 10280e53-spkid0 firmware, which upstream never published because only spkid1 to spkid3 exist. That is issue #9687, reproduced in #10543, #10987 and on two units in #11320, which its reporter closed as a duplicate of #9687. On SKU 0E54 there are no 10280e54 firmware links at all, per #11356. Symptoms in the kernel log are FIRMWARE_MISSING, Calibration disabled due to missing firmware controls and Can't read tuning IDs.

The tracked upstream bug is kernel.org bugzilla 221956, linked from #9687. Cirrus traced the spkid0 reading to a driver bug: the ACPI GPIO resource for the speaker ID has two pins and spi-cs42l43 only read one. With Richard Fitzgerald’s patch the amps resolve spkid2, load the published tuning and log Calibration applied. Four people have confirmed it on 0E53 units, wizaj, sg-qwt and coldSpidey in #9687 and Gundrak on two machines in #11320, but every one of them built it as an out-of-tree module, which a kernel update silently reverts. Nitad’s follow-up in #11374 after moving to 4.0.4 confirms linux-omarchy 7.2.5-3 does not carry it. wizaj has withdrawn the earlier recipe of symlinking spkid0 onto the spkid1 files and asks that nobody use it, since the tuning is speaker-specific. For 0E54, #11356 added the nine 10280e54 links listed in the upstream linux-firmware manifest and got working speakers on a unit that reads spkid1, while Nitad’s 0E54 unit also needs a driver change because its ACPI tables do not describe the speaker-ID GPIOs at all.

If you just want audible sound while this is open, two reporters got the tweeters back by disabling the sidecar route: a1local in #9687 with options snd_soc_sof_sdw quirk=0 and phuclh in #10987 with quirk=1. That is the pre-#7427 state, thin and bass-less, but it plays.

Sound card fails to appear on some boots. Issue #10987 reports roughly half of boots ending in sof_sdw: ASoC: failed to instantiate card -16, alternating between the left and right amp, leaving PipeWire with Dummy Output only. The reporter’s reading is a probe race between the two SPI amps. Unbinding and rebinding sof_sdw recovers the card for that boot, per a1local in #9687.

Panel Replay judder. The 120 Hz panel negotiates Panel Replay Selective Update, and its wake path drops frames in multiples of the refresh period. Scrolling and the mouse cursor stutter. Issue #6853 has the diagnosis and the one line fix, xe.enable_psr=0 xe.enable_panel_replay=0. A later comment there reports xe.enable_psr=2 xe.enable_panel_replay=0 also works with less battery cost. The pull request that would ship it, #6849, was still open when this page was checked, so 4.0.4 does not apply it for you.

Suspend on the 4.0.4 kernel. Version 4.0.4 makes linux-omarchy the default boot entry for everyone. Issue #12190 reports linux-omarchy 7.2.5-3 wedging on lid close on this model, backlight on and input dead, four forced power cycles in a day, while the stock kernel on the same command line suspends fine. It is one reporter so far, filed on 2026-09-16.

Wi-Fi on the Intel BE213 card. Issue #6551, filed on an XPS 13 that the reporter does not identify beyond the BE213 card, shows iwlwifi finding no acceptable bz-b0-wh-b0 ucode because the install media carries linux-firmware 20260410-1. With no wired port, that is a fresh install with zero network. The reporter fixed it by sideloading linux-firmware-intel and a newer kernel with pacman -U. Note that a normal update left linux-firmware-intel behind on the old version.

What Omarchy does for this model

Match strings come from omarchy-hw-match, which greps /sys/class/dmi/id/product_name and product_family case-insensitively.

• omarchy-hw-dell-xps-haptic-touchpad matches XPS plus the presence of /sys/bus/i2c/devices/i2c-VEN_06CB:00, and install/hardware/dell-xps-touchpad-haptics.sh then installs dell-xps-touchpad-haptics, version 1.0.0-3 in the stable channel.
• install/user/hardware/dell/xps13-text-scaling.sh matches DX13260 and sets the text size.
• omarchy-hw-dell-xps-oled requires Panther Lake graphics and an LG EDID manufacturer id, but no install script in 4.0.4 calls it, so it changes nothing on any XPS today.
• install/hardware/intel/fix-wifi7-eht.sh disables Wi-Fi 7 only for PCI ids 8086:e440 and 8086:272b, the BE200 and BE211 in the XPS 14 and 16. The BE213 in the XPS 13 is 8086:4d40 and is not covered.
• The shipped speaker tuning dell-xps-2026 matches DMI SKUs 0DB9 and 0DBA, which are the XPS 14 and XPS 16. There is no tuning for the XPS 13. Check yours with omarchy audio tuning status.
• dell-xps13-sidecar-amps exists in the Omarchy repo at 1.0.0-2, but no install script adds it. On kernel 7.2 and newer it is a no-op, because upstream commit efd80de2de9d already sets the sidecar quirk for 1028:0e53.

Variants

The DX13260 ships with either a Core 5 320 on Wildcat Lake (#6853, #12190) or a Core Ultra 7 355 on Panther Lake (#11356, #11374), and in at least two speaker SKUs, 0E53 and 0E54. Neither SKU is currently safe for speaker audio. Read your own with cat /sys/class/dmi/id/product_sku.

Older XPS 13 models bring their own problems. The XPS 13 Plus 9315 is Alder Lake with an IPU6 camera, and Omarchy’s camera detection only covers IPU7, so nothing is installed and the webcam stays dark, per #9879. The Lunar Lake XPS 13 9350 gets intel-ipu7-camera by detection, but that package is built for Panther Lake only, so the HAL plugin its IPU needs does not exist, per the reporter of #12178. Goodix fingerprint readers with no open libfprint driver turn up in several Dell laptops, including an XPS 13 9310 in a comment on #7991. We have no confirmed fingerprint report for the DX13260 either way.

If you are buying new for Omarchy, the XPS 14 and 16 have the speaker tuning, the Wi-Fi 7 workaround and the camera enablement that the 13 does not.

Before you install

• Have a USB Ethernet adapter or phone tethering ready in case Wi-Fi does not come up.
• Record your SKU and audio subsystem id before you change anything: cat /sys/class/dmi/id/product_sku and lspci -nn | grep -i audio.
• After the first boot, run journalctl -b -k | grep cs35l56 and look for FIRMWARE_MISSING. The system name: line that tells you which spkid your unit asks for only appears once the amps bind into the card, per a comment on #10543.
• Expect to keep the stock linux entry available in the boot menu while #12190 is open. See system sleep in the manual.
• If scrolling judders, try the Panel Replay kernel parameters from #6853 yourself, since Omarchy does not apply them.
• All of the above was checked against 4.0.4 on the 4.x line. None of these issues predate Quattro in any report we found. The 3.x-era XPS 13 reports, such as #1868 on a 9315, are older units with unrelated audio routing problems.

Related

Component pages: audio, wifi, suspend and sleep, webcam, touchpad and input, intel gpu. Fix pages: no sound from laptop speakers, suspend will not resume, webcam not detected. Sibling model: Dell XPS 14. Report your own unit at hardware submit.