[satchlj]

Keep the exact J613 MTP bypass stream alive for live RTKit adoption, with identity-domain attachment and readback checks. Retain Aurora’s independent J614s/J713, radio and display policies. Requires the matching loader in aurora-silicon/u-boot#30.

Port of omacom/linux-aurora#328. The original PR remains untouched.

Validation: arm64 defconfig + asahi.config, Image/modules/DTBs with BTF and KUnit enabled, LLVM 18/Rust 1.85.0, passed at stack top #343 (11d802aa9a17), which includes this branch. The validated integration base is b500668. QEMU loaded the test-bearing modules successfully with no additional failures. The combined run retains four pre-existing SEP store failures (plus their suite aggregate), reproduced on the unmodified Aurora base. The unmodified pinned base also crashes in apple-av-audio-lifetime/av_lifetime_devres_order; that unrelated audio suite is explicitly excluded. All tests added by these ports run through KUnit debugfs on Cortex-A76. This is not a fully green or complete-tree KUnit run. No new host fixture was authored for this layer; the arm64 module build and QEMU module-load smoke test passed. No 7.3 hardware boot or deployment has been performed.

[satchlj]

@DjDeveloperr @RyanTheTide ready for your regression testing. It hasn't been boot-tested on J613 with 7.3 yet; I'll post the result here when I have it.

[satchlj]

J613 boot result on 7.3 (rubra, M3 MacBook Air 13", macOS 26.6.2 / 25G83 firmware), 2026-11-05.

• Kernel: integration/aurora 11eccbed with all 17 J613 PRs merged (soc: apple: sep: preserve unreadable and partial identity state #962–media: apple: isp: bound exposure metadata and keep controls gated #978), plus HID: dockchannel-hid: write each frame to the FIFO under one lock #1010 (the MTP frame-lock fix). Boot chain: stage2 from the m1n1 Thunderbolt display on M1 Pro/Max (t600x), stacked on #8 #64 content, U-Boot with arm64: dts: apple: t8140/j700: keyboard, trackpad and USB-C topology (reworked, 5 patches) #30/soc: apple: sep: T6000 (M1 Pro) support, plus SPI CS timing and attach fixes #26, then Limine.
• Mesa: aurora-wip 4371cf86 + mesa cpufreq: apple: add bounded M6 T8152 CPU scaling #161–#163.
• Automatic checks passed:
  • boots to the desktop;
  • GPU: DRM driver aurora, Apple M3 (G15G) for GL and Vulkan, glmark2 1113 (1011 on 7.2.11), no GPU errors;
  • internal panel up;
  • keyboard and trackpad present with no MTP crash;
  • camera node, Wi-Fi, Bluetooth, audio cards and headphone jack;
  • Touch ID policy readback (identity_source=dt);
  • speakers stay off as intended (our stage2 passes no apple,speaker-config).
• Not checked yet, since they need me at the machine: the fingerprint match, an external display, and suspend.

[RyanTheTide]

The new probe check rejects any DART that carries apple,bypass-handoff-streams unless it is the J613 instance. On t8140 (J700) the keyboard/trackpad DART at 314800000 has that property with value 0, so it now fails to probe:

apple-dart 314800000.iommu: probe with driver apple-dart failed with error -22

The MTP then only appears ~20 s into boot after the deferred-probe timeout, without its DART. Please restrict the new path to the boards it is meant for (or treat a zero/absent stream mask as "no handoff" and fall through to the existing probe) so other SoCs keep their current behaviour.

[satchlj]

Rebased onto current integration and fixed the zero-value case: J700's apple,bypass-handoff-streams = <0> now takes the existing probe path, as an absent property does. Nonzero handoffs still require the qualified J613 instance. The actual admission block passes offline absent/zero/nonzero/J613 checks. I have not boot-tested this new head on J700 or J613.

[satchlj]

Claude Opus 5.5 on @satchlj's behalf

Tested on J613 (MacBook Air M3), system firmware 26.6.2 / 25G83. Kernel: exact head
f891a957fc26e8c9e6b1823f25e1cfd4f37b2c59 with no extra commits, release
7.3.0-aurora73-pr971-f891a95-ARCH (packaged default config, debug info/BTF off), UKI
sha256 38788238…3f48f01, booted as a one-shot trial.

• rtkit-helper 2fa400000.mtp: adopted the running RTKit session (4 retained buffers).
  dockchannel-hid interfaces stm, keyboard, actuator and multi-touch all ready. Input
  devices: "Apple MTP keyboard", "Apple MTP multi-touch", "Apple MTP multi-touch Force
  Click"; MTP haptics initialised.
• Physical input during a 60 s evdev capture: trackpad 26,933 ABS events and 33 button
  presses; keyboard 9 key events from 2 distinct keys. The person at the machine
  confirmed that typing, pointer movement and clicks behaved normally.
• No Oops, BUG, call trace, DART fault or MTP crash in dmesg.
• Baseline for comparison: an upstream integration/aurora kernel without this PR on
  the same machine had no MTP driver bound and no keyboard or trackpad.
• Not tested: force-click and haptic feedback deliberately, suspend/resume of MTP, long
  uptime.

This also means a J700 boot of this head is still owed for the zero-stream fix; I don't have one.

[satchlj]

GPT-6.1 Sol on @satchlj's behalf

2026-11-10 hardware re-test with #971 head 6debb6f. Matched running 30.6 devices/platform PM tests reached their five-second checkpoints. The previous f891a95 composition aborted at apple-dart 2fa808000.iommu with -16. Its sysfs consumers are MTP 2fa400000 and input 2fab28000. With the new suspend/resume bypass guards, freezer/devices/platform each returned rc=0 and reached all three five-second checkpoints. No DART EBUSY or retained-state-loss message; keyboard/multitouch/Force Click devices remained registered afterward. pm_test restored to none. No RTC wakealarm was available; no real s2idle, wake or sleep-power test was run.

J613/T8122 M3 Air, integration dc8f6f7 plus #971/#973/#974/#978/#1340/#1486/#1491, composed source 112007b70271. Hardware-tested release 7.3.0-aurora73-upstream-2-ARCH; UKI SHA256 248283ad7750793cf60f8530e3e2331f7c1a58d733d2f3e8880f8a959e87becc. Composition evidence, not an isolated PR kernel. The same binary passed a one-shot trial and the normal new-default reboot. Installed package revision 30.10 corrects only the installer's pre-extraction space estimate; all 2054 Linux-package usr regular files match the qualified 30.9 package. Stage1/Stage2 unchanged, four boot hooks masked.

[DjDeveloperr]

Codex (GPT-6.1 Sol), reviewing on @DjDeveloperr’s behalf.

No blocking findings on the current head. The admission check now treats absent/zero bypass-handoff streams as no live handoff, preserving the existing J700 path; nonzero handoff remains restricted to the qualified J613 DART/stream with bypass readback. Latest J613 suspend/resume guards retain the live epoch and refuse deeper sleep. Exact-head source, Rust, KUnit and host checks pass. Posted J613 physical input and later composition PM diagnostics are attributed evidence, not fresh tests by this review. A J700 zero-stream boot/input smoke and deliberate force-click/long-uptime qualification remain rollout follow-ups; real MTP retention for the distinct #972 implementation is not inferred from these results.