MacBook Pro · M2 family

[Acelogic]

MacBook Pro · M2 family

🧭 Hardware directory › M2 category › MacBook Pro

5 board configurations · one shared discussion · Aurora Linux assessment

Broad kernel integration is present: Aurora's working tree includes this product's core platform, NVMe, GPU, DCP display, USB, wireless and audio paths. Board-specific exceptions and Aurora's newer development work are listed below; this assessment is not a blanket pass for every port, firmware version or installation.

Refreshed October 16, 2026. Baseline: aurora-wip at f53454ba0da1. Open PR and feat/sep results are identified separately. These statuses use Aurora's public source and contributor reports with their actual tested kernels identified. The baseline source and every linked PR state were rechecked; no new physical test was performed for this update.

Configurations

Board	Silicon / SoC	Model / configuration	Year	Aurora assessment
J493	M2 · T8112	MacBook Pro 13-inch	2022	🔵 Broad kernel integration
J414S	M2 Pro · T6020	MacBook Pro 14-inch	2023	🔵 Broad kernel integration; board reports below
J416S	M2 Pro · T6020	MacBook Pro 16-inch	2023	🔵 Broad kernel integration; board reports below
J414C	M2 Max · T6021	MacBook Pro 14-inch	2023	🔵 Broad kernel integration; board reports below
J416C	M2 Max · T6021	MacBook Pro 16-inch	2023	🔵 Broad kernel integration; board reports below

Evidence key · Aurora source and Aurora hardware reports

Marker	Meaning
🔵 Integrated source	The relevant implementation or enabled board description exists in the pinned Aurora working tree. This is source evidence, not a new hardware test.
🟢 Hardware report	A linked public Aurora report names the board and tested feature. Read its revision and limits.
🟠 Development / partial	An open PR, feature branch, incomplete configuration or qualification remains.
🔴 Source gap	The required driver match or board path is absent or disabled in this snapshot.
⚪ Unverified	The evidence checked here does not establish this capability.
—	Not fitted to this product.

A report from another board is not a pass for this board. An open PR may work on hardware without being integrated into aurora-wip. A driver being enabled does not by itself establish firmware availability, userspace integration or a complete usable installation.

🧩 Core platform

Feature	Aurora status	Evidence / boundary
Board description / kernel build target	🔵 Integrated source	Every board above has a DTS and a Makefile target. Bootloader handoff and a repeatable installer run are separate checks.
NVMe / ANS storage	🔵 Integrated source	Enabled storage nodes in every board tree; CONFIG_NVME_APPLE=m.
CPU frequency / idle	🔵 Integrated source	Enabled cluster-frequency nodes; Apple cpufreq, CPU idle and PMU options configured. CPU idle is not whole-system suspend qualification.
Interrupts / IOMMU / platform buses	🔵 Integrated source	Apple AIC, DART, GPIO, I²C, SPI, SPMI and RTKit platform options are configured; board-specific nodes are in the linked DTS sources.
SMC / RTC / sensors	🔵 Integrated source	SMC paths present; MACSMC RTC, input, hwmon and power drivers configured.

🎨 Graphics, audio and media

Feature	Aurora status	Evidence / boundary
GPU acceleration	🔵 Kernel driver integrated	Enabled AGX node and matching driver entry for the listed SoCs; CONFIG_DRM_ASAHI=m. Mesa, firmware and application / conformance results must be recorded separately.
Display pipeline	🔵 DCP integrated; 🟢 J416S DCP audio and Type-C reports	Enabled DCP / DCPEXT paths and CONFIG_DRM_APPLE=m. Merged linux #282 answers the DPTX port's group 1 calls as macOS does; on J416S under the m1n1 hypervisor it removed the first-audio-start refusals (reports below). Merged linux #224 (legacy reply records) and linux #225 (released register ranges) fix the pre-H17P DCP paths that M1/M2 firmware uses (tested on J413 and J316S); merged linux #313 was qualified on J700 only. DCP audio fixes linux #276, linux #277 and linux #299 remain open. Connector routing, refresh modes, brightness and hotplug remain board- and firmware-specific.
Speakers / headphone jack	🔵 Speaker + headphone links	MCA / Mac audio drivers and codecs are configured. Audio routing in source does not establish safe speaker tuning or a playback qualification.
Built-in microphones	🔵 AOP audio enabled	Board AOP audio state checked; CONFIG_SND_SOC_APPLE_AOP_AUDIO=m. Record actual capture quality and channel behaviour per board.
Built-in camera	🔵 ISP integrated	Enabled ISP node with a matching driver family and CONFIG_VIDEO_APPLE_ISP=m. Camera sensor / firmware and runtime capture still need board evidence.
Hardware video decode	🔵 AVD source; 🟠 qualification	AVD nodes and decoder driver are present and CONFIG_VIDEO_APPLE_AVD=m. Decoder fixes in #45 are merged into aurora-wip; codec / stream coverage and application integration are not a general pass.
Video encode / ProRes	⚪ Unverified	This audit does not establish an integrated, hardware-qualified encode or ProRes pipeline.
Neural Engine	🔵 Driver + runtime PM; chip-specific enablement	linux #155 and buffer/lifetime fix linux #191 are merged. Runtime PM autosuspends after 1.5 s by default. The pinned DT enables T8103, T6000/T6001 and T6020/T6021; base M2/T8112 and M1/M2 Ultra remain disabled. Firmware and userspace are separate requirements. Since October 12, asahi.config selects CONFIG_DRM_ACCEL=y and CONFIG_DRM_ACCEL_ANE=m (direct merge 45da41c0cd1b); the arm64 defconfig alone does not, and a configured driver is not a per-board runtime test. J293 and J314S each completed 600/600 bit-exact add calls over 30 autosuspend/resume cycles, four concurrent 20-call runs and a device-test suspend cycle on contributor kernel 11.31/custom/sep@87e165cc (report). This is not a full-system suspend test or a direct power measurement. A report records passing in-tree smoke rows for M1, M1 Pro and M2 Pro; chip enablement does not qualify every product.

🔌 I/O and security

Feature	Aurora status	Evidence / boundary
USB2 / USB3 host	🔵 DWC3 integrated	Enabled Apple DWC3 board nodes and configured driver. Port combinations, hubs and role swaps need per-board tests.
USB4 / Thunderbolt / USB-C display	🔵 Driver configured; 🟠 board-specific qualification	linux #8, linux #50, linux #184, linux #193 and linux #195 are merged and their merge history reaches the pinned tree. asahi.config selects CONFIG_USB4=m and CONFIG_USB4_APPLE_SOC=m. linux #46 and linux #51 closed without merge on October 11; #8 had already carried their dual-stream and controller-teardown lock changes, as it carried linux #64. Merged linux #327 lets a CD321x port register without its USB4 router once that router fails to bind (builds without USB4_APPLE_SOC); it was tested on J316S and J413 only. J416S Type-C and tunnel results on aurora-wip under the m1n1 hypervisor (merged linux #293, open linux #296) and a J416S report on a non-Aurora tree are in the reports below. A default-on T600x/T602x PCIe-C cold-init path (pcie_apple.tunnel_kernel_init) and the T602x PCIe-C DART tunables reached the pinned tree on October 12 through direct merge 45da41c0cd1b. Display, PCIe tunnelling and resume remain board/build specific.
Wi-Fi / Bluetooth	🔵 Wi-Fi board path; 🔵 Bluetooth board path	Broadcom recovery linux #7, AWDL linux #24, J416C recovery linux #167, recovery teardown linux #186 and failed-reset handling linux #192 are merged. Firmware, userspace and named-board test limits remain; this does not qualify every wireless function.
Keyboard / touchpad	🔵 Built-in input path	SPI HID or dockchannel HID descriptions and Apple input drivers are present. Basic input and trackpad haptic feedback are distinct.
Ethernet / SD / PCIe expansion	⚪ Per-device qualification	PCIe host support is configured. Qualify fitted controllers, adapters and link speeds independently; do not copy support from a sibling board.
SEP / Touch ID	🔵 Integrated driver and board nodes; 🔴 no J493 sensor node; 🟠 board-specific acceptance	SEP integration is present through merged linux #42. The pinned tree has the T6020 SEP profile (T6021 uses it) and enables the Mesa sensor on J414S, J416S, J414C and J416C through the shared board file; J493's Mesa node stays disabled. linux #69 and linux #152 themselves are merged only into feat/sep. Merged linux #303, linux #325 and linux #326 are J700 SEP fixes (#325 and #326 state that M1/M2 are unaffected); merged tools linux #329 and linux #330 add the Fedora SELinux label and PAM step. J414C has a new contributor daily-use report on an omacom/linux-aurora release (below). Named sensor, firmware and tested-kernel limits still apply; downstream iconidentify/aurora-linux fixes are separate from this snapshot.

🔋 Power and complete-system readiness

Feature	Aurora status	Evidence / boundary
Battery / charging / thermals	🔵 Driver foundation	MACSMC power and thermal / hwmon options are configured. Charge behaviour, runtime and battery-health reporting need a board test.
Suspend / resume	🟠 Board-specific qualification	CPU idle and shutdown drivers do not prove a complete suspend / resume cycle. Only the exact boards and development builds named below have cited runtime results. J416C now has an undocked s2idle display-wake report on merged #8; Bluetooth can fail after some resumes on that board. J416S passed pm_test=platform s2idle 6/6 under the m1n1 hypervisor on merged #293's head, the same as its revert control; native J416S suspend on aurora-wip is untested.
Installer / complete system	⚪ Release qualification unverified	This is a public kernel source and evidence audit, not a fresh installation or end-to-end hardware test. Record installer, boot chain, kernel, Mesa and firmware revisions together.

🧪 Aurora hardware reports and development branches

Board	Feature	State / source	Result and limits
J414C	Touch ID after reboot	🟢 Reported · 🟣 merged #69 → feat/sep	3/3 verification matches after reboot on commit-3 / test-11.5 work. Earlier enrolment work in #58 merged into feat/sep; current SEP source is integrated through #42. That integration does not rerun this historical hardware result.
J414S	Touch ID enrol / verify and PAM sudo	🟢 Separate report · 🟠 closed without merge #49	Historical J414S profile report. A later report on contributor 11.32 plus iconidentify/aurora-linux#17 (enrolment capture budget raised from 12 to 36) records enrol/verify, cold-reboot verification, one-finger deletion persistence and verification after s2idle on firmware 26.0 / OS firmware 13.5; those downstream fixes are not claimed for pinned aurora-wip. J414C evidence does not qualify J414S.
J416S	USB peripherals through a Thunderbolt hub	🟢 Reported · 🔵 merged #6 → aurora-wip	OWC Thunderbolt 5 hub with USB keyboard. This result does not qualify DisplayPort or PCIe tunnels.
J416S	Thunderbolt display / resume work	🟢 Board tested · 🟣 merged #8	This is the tested T602x board in #8. Code also covers other Pro / Max laptops, but those boards do not inherit this result.
J414S / J416C	Dual-dock display work	⚫ linux #46 closed without merge (October 11); 🟣 merged linux #50 and linux #8	#8 carries #46's dual-stream commits and part of #50; #46 and its follow-up linux #51 closed after that. A T600x/T602x PCIe-C cold-init path later reached pinned aurora-wip through direct merge 45da41c0cd1b. A J416C report on #8 head 581d7e36ea84 runs two dock displays on dcpext0/dcpext1 (row below). That run used a dock, so #8's direct DP-alt compositor-pairing fix is still untested on J414/J416.
J414C	Neural Engine research	🟠 Historical linux #65; newer linux #155 merged	The earlier external setup returned intermittent zero output. #155 reports a later T6021 in-tree driver with three bit-exact whole-encoder runs and firmware loading, but does not identify the board and used a different boot-image DT. Do not transfer that result to every M2 Max board.
J414S	Bluetooth reset recovery	🟢 Contributor report · 🟣 merged linux #7	Mac14,9 / M2 Pro report exercised teardown failure, reset recovery, firmware restoration and paired audio/input reconnection on its named 7.1.13-3-1 kernel; not a current-baseline retest.
J416C	Thunderbolt 3 dock displays, hotplug, clamshell and s2idle	🟢 Contributor report · 🟣 merged linux #8	#8 head 581d7e36ea84 plus two local patches, m1n1 1.6.1, CalDigit TB3 dock. Two 3840×2160@60 dock monitors plus the internal panel at boot; displays returned on every repeated hotplug; docked lid-close stays awake; all three outputs return after undocked s2idle. Each attach's first PCI tunnel activation fails and succeeds on retry. No wake from a dock USB keyboard.
J416C	PCIe tunnel devices: dock USB, Ethernet, SD reader, webcam	🟢 Contributor report · 🔵 cold-init path now in pinned aurora-wip (direct merge, no PR)	Worked, including replug and replug after s2idle, only with two patches on a contributor branch. They replace #50's cold init with #8's: PCIe-C kernel-init DT and reset lines for J414/J416, and t600x/t602x cold init behind pcie_apple.tunnel_kernel_init=1. Since October 12, pinned aurora-wip has a T600x/T602x cold-init path under the same parameter, on by default, plus T602x PCIe-C DART tunables (direct merge 45da41c0cd1b); the J416C run has not been repeated on it. Tested only on J416C with one TB3 dock; other J414/J416 boards do not inherit it.
J416C	Bluetooth recovery after s2idle	🟢 Contributor report · 🟣 merged linux #167	BCM4388 stops answering HCI after some resumes (-110 timeouts until a module reload). #167 reprobes the driver on a command timeout. On 7.1.12 (#8 plus the patch), an MX Master 3S reconnected after several s2idle cycles. No log yet shows the handler firing on a real failure. It recovers the controller and does not fix the root cause.
J414C / J414S / J416C / J416S	Shared Touch ID sensor description	🔵 Source in merged linux #69	October 2 follow-up: shared J414/J416 wiring added; J414C DTB unchanged and end-to-end operation reported. J416C lines were confirmed, but J416C/J416S runtime was untested in that report.
J414C	Touch ID in daily use: sudo, polkit, enrolment	🟢 Contributor report on an omacom/linux-aurora release	Arch Linux ARM; release aurora-2026.10.13.18, kernel 7.1.12-2-20261013.18-sep-ARCH built from 6f6367775185 (a downstream merge that has diverged from pinned aurora-wip), m1n1 1.6.1-omarchy.aurora17, aurora-touchid 20261003-2, libfprint 1.94.100 with Apple SEP support, fprintd 1.94.5, OS firmware 13.5, system firmware 27.0. Reported working in normal use; no repeat counts. It does not qualify J414S or J416x. The report also mentions a Surface USB4 Dock with two external monitors, without details.
J416S	DP over USB4 with DSC, two displays, hub, Ethernet, suspend	🟢 Contributor report on a non-Aurora tree	NixOS on the Asahi fairydust branch with a USB4 backport (tree, HACK commits included): a Dell U2725QE over Thunderbolt with mode switches and DSC up to 4K@120, every USB-C port, two displays, the monitor's hub and Ethernet, and suspend/resume with the monitor attached, in daily use. DP/HDMI audio was still in progress. No repeat counts or exact revision; not evidence for pinned aurora-wip. A follow-up says aurora-wip boots under the m1n1 hypervisor with a bench-only DT change, which the next row uses.
J416S	DCP audio and Type-C replugs on aurora-wip under the m1n1 hypervisor	🟢 PR reports · 🟣 merged linux #282, linux #293; 🟠 open linux #276, linux #277, linux #296, linux #299	DCP firmware 13.5, CONFIG_RUST=n, and a bench-only DT change that keeps atc0 (held by m1n1) out of the DCP routes; a TV on a USB-C to HDMI adapter (dcpext0) and a Dell U2725QE over Thunderbolt (dcpext1). #276: the TV's 17312-byte audio element list is read in full and a failed START no longer underflows runtime PM. #282: first audio prepare accepted in 5/5 boots (refused 3/3 without). #299: 16/16 TV replugs reached the display without exhausting AV service slots; 95/120 starts played in full, the rest around replugs. #296: 0/24 lost TV replugs (4/16 without). #293 run: Thunderbolt tunnel 8/8 and DP-alt TV 8/8 replugs with audio playing, pm_test=platform s2idle 6/6, matching the revert control. Native boots, ATC0 and full suspend are untested; open PRs are not in pinned aurora-wip.

PR snapshot: October 16, 2026. Current states and destinations are recorded below. SEP is integrated through #42, while newer downstream Touch ID fixes retain their contributor-kernel scope.

What changed · October 16, 2026

• J414C Touch ID: a contributor report on omacom/linux-aurora release aurora-2026.10.13.18 has sudo, polkit and enrolment working in daily use. It does not qualify J414S or J416x.
• J416S on a non-Aurora tree: discussion #269 reports DP over USB4 with DSC up to 4K@120, two displays, the monitor's hub and Ethernet, and suspend on Asahi fairydust with a USB4 backport.
• J416S on aurora-wip under the m1n1 hypervisor: merged linux #282 and linux #293 and open linux #276, linux #277, linux #296 and linux #299 carry DCP audio and Type-C replug results, using a bench-only DT change.
• Thunderbolt source: linux #46 and linux #51 closed without merge after Thunderbolt: M1/M2 Pro display, M1 PCIe and resume recovery #8 carried their changes. A T600x/T602x PCIe-C cold-init path with T602x DART tunables arrived through direct merge 45da41c0cd1b.
• Shared rows: asahi.config now selects the ANE driver; the SEP row records that J493 has no Mesa sensor node enabled.

Pinned Aurora evidence

Working-tree configuration · Arm64 base configuration · Board build targets · GPU device matches · DCP driver · ISP driver · AVD decoder · Thunderbolt driver · Thunderbolt configuration · SEP driver

Public references

• The Apple Wiki board identifiers — public identity cross-reference.
• M2 series device catalogue — identity source, not a support guarantee.
• Linux source snapshot — aurora-wip at f53454ba0da1.

Board-specific identity and device-tree references

• J493: Device tree.
• J414S: Device tree.
• J416S: Device tree.
• J414C: Device tree.
• J416C: Device tree.

Related board work

• J414S: #46 (closed without merge on October 11; carried by Thunderbolt: M1/M2 Pro display, M1 PCIe and resume recovery #8) — drm/apple: support independent dual dock DP streams on J414s
• J416S: #6 (merged into the PR's target branch; see branch-qualified evidence above) — j416s: USB4 hub USB (CD321x USB4-before-DP + m1n1 1.6.1 aliases)
• J414C: #58 (merged into the PR's target branch; see branch-qualified evidence above) — soc: apple: sep + dts: Touch ID on the M2 Max (t6021)
• J416C: #50 (merged) — J416c: dual dock DP streams, DP connector type, and PCIe-C for Thunderbolt 3 docks (stacked on drm/apple: support independent dual dock DP streams on J414s #46)
• J416C: #167 (merged) — Bluetooth: hci_bcm4377: Rebind the driver when the controller stops responding
• J416S: #276, #296 and #299 (open) — DCP audio elements and runtime PM, CD321x empty-port sleep, AV service slot reuse; #282 (merged into aurora-wip) — DPTX group 1 calls
• J414C: Touch ID report (contributor, omacom/linux-aurora release) — this discussion

Open and draft PRs are development work; a merged change alone does not qualify every configuration above.

🧪 Add a reproducible hardware report

Reply here with one feature or failure per report. Identify the exact row from the configurations table.

Board ID:
SoC ID and chip variant:
Model / screen size / configuration:
Feature or SoC block:
Kernel repository / branch / commit:
m1n1 / U-Boot / Mesa commits, where relevant:
Firmware version:
Steps to reproduce:
Expected result:
Observed result and repeat count:
Public log or artifact URL:
Artifact SHA-256 and command to repeat the check:
Known limits / recovery steps:

Repository admins can update this opening post through ⋯ → Edit. Record the tested board next to each result; keep implementation status separate from unverified hardware results on other boards.

---

Keep this product generation together. Discuss shared silicon changes here and link affected sibling products through the hardware directory. A result on one board does not change another board’s status.

Audit baseline: f53454ba0da1; public source and PR reports checked October 16, 2026. No new physical test was performed for this update. Working-tree integration does not imply an aurora-stable release.

PR state and integration audit · October 16, 2026

PR	GitHub state	Destination / target branch	Integration evidence
linux #6	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · 9220af926e67
linux #7	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · e6c5b30d0ac0
linux #8	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · b37df4bdd552
linux #24	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · 6c00439340a7
linux #42	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · c9a55efdff8f
linux #45	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · 510ac224fe96
linux #46	⚫ Closed without merge	review/pr14-base	Closed without merge; any carried changes must be checked independently · 2530a2c3a694
linux #49	⚫ Closed without merge	feat/sep	Closed without merge; any carried changes must be checked independently · 8387e8ac148d
linux #50	🟣 Merged	fix/j414s-dual-dock-dp	Merge is in the pinned aurora-wip history · 9dc79fb25c92
linux #51	⚫ Closed without merge	fix/j414s-dual-dock-dp	Closed without merge; any carried changes must be checked independently · 8153565e7166
linux #58	🟣 Merged	feat/sep	Merge is in the pinned aurora-wip history · 74f6fe49cbee
linux #64	⚫ Closed without merge	aurora-wip	Closed without merge; any carried changes must be checked independently · 5d524a478c92
linux #65	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · 2e1656f93c27
linux #69	🟣 Merged	feat/sep	Merged to feat/sep; SEP integration also landed through #42. Later downstream fixes are not implied · 2cf2b0a22229
linux #152	🟣 Merged	feat/sep	Merged to feat/sep; SEP integration also landed through #42. Later downstream fixes are not implied · 120e030eac9f
linux #155	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · 9e9e1b83b71e
linux #167	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · 883e58b33d57
linux #184	🟣 Merged	fix/j414s-dual-dock-dp	Merge is in the pinned aurora-wip history · 9e93cac3ac71
linux #186	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · 0fc050327e0c
linux #191	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · 212a52e404db
linux #192	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · 49d3482292fd
linux #193	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · b1bd8bd99cf0
linux #195	🟣 Merged	fix/j414s-dual-dock-dp	Merge is in the pinned aurora-wip history · 2530a2c3a694
linux #224	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · 0a0216f71a14
linux #225	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · 98818cea22ee
linux #276	🟠 Open	aurora-wip	Open development PR; hardware reports retain their tested revisions · e2230a4db1b8
linux #277	🟠 Open	aurora-wip	Open development PR; hardware reports retain their tested revisions · 7b6a4a7c42ae
linux #282	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · e7fe93c5f765
linux #293	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · d588f760941e
linux #296	🟠 Open	aurora-wip	Open development PR; hardware reports retain their tested revisions · d1c7924cf203
linux #299	🟠 Open	aurora-wip	Open development PR; hardware reports retain their tested revisions · 1a6694d5fb56
linux #303	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · a490adab6fc3
linux #313	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · 5cf6579fa592
linux #325	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · 97075ac83279
linux #326	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · 99b5f912426e
linux #327	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · a6cfa4dd59a5
linux #329	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · 8b66c7bc0ec6
linux #330	🟣 Merged	aurora-wip	Merge is in the pinned aurora-wip history · a837bd5f9a21

PR state, source integration and hardware qualification are separate. A closed PR is not necessarily merged. Public reports describe their original test kernels, including contributor branches. Repository administrators can edit this opening post.

[muness]

J414C / 14-inch M2 Max: working Touch ID on Arch Linux ARM

Reporting working built-in Touch ID on my 2023 14-inch M2 Max MacBook Pro, running Arch Linux ARM with Aurora’s published kernel and Touch ID packages.

Field	Value
Board ID	apple,j414c
SoC / chip	apple,t6021 / M2 Max
Model	MacBook Pro, 14-inch, 2023
Operating system	Arch Linux ARM, rolling, aarch64
Feature	Built-in Touch ID fingerprint authentication
Aurora release	aurora-2026.10.13.18
Running kernel	7.1.12-2-20261013.18-sep-ARCH
Kernel package	linux-aurora 7.1.12.aurora2-20261013.18
Published kernel source commit	6f63677751854d57bcfac172314957bc2197c05f
m1n1 stage 1	v1.6.1
m1n1 stage 2	v1.6.1-omarchy.aurora17
m1n1 package	m1n1-aurora 1.6.1.aurora17-5
U-Boot	Runtime 2026.07; package uboot-asahi 2026.07.asahi2-1
OS firmware	13.5
System firmware	27.0
Touch ID package	aurora-touchid 20261003-2
Fingerprint library	libfprint 1.94.100-1.1, with Apple SEP support
Fingerprint daemon	fprintd 1.94.5-2

Observed result: Touch ID is working in normal use on this machine.
Authentication contexts confirmed working: sudo, polkit, Enroll

Supporting observations: The Apple SEP service starts successfully. Fingerprint PAM entries are configured for sudo, polkit, and the Omarchy fingerprint lock helper.

---

I also have successfully used the laptop with Surface USB 4 Dock, including 2x external monitors. Happy to post details if that helps.

[muness]

Update: upgraded to aurora-2026.10.17.20 (7.1.12-2-20261017.20-sep-ARCH).

• Touch ID still works after reboot with existing enrollment; coordinated fprintd-verify returned verify-match.
• Surface Thunderbolt 4 Dock: Dell U3818DW at 3840×1600/60 Hz and LG 4K at 3840×2160/60 Hz, plus the internal panel at 120 Hz. Both externals recovered after one unplug/replug.
• Separately, trying apple_pmgr_misc.idle_fabric_pstate=1 while undocked was followed ~18 seconds later by a display freeze and DRM page-flip errors. Required a forced restart. One attempt only; governor is back at 0. Logs and trial data available.