Repository navigation
Conversation
|
@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. Still a draft until it's tested on a J615. |
|
J613 boot result on 7.3 (rubra, M3 MacBook Air 13", macOS 26.6.2 / 25G83 firmware), 2026-11-05.
For this PR: the J615 code is built in and stays inert on J613. It still needs a J615 to test. |
RyanTheTide
left a comment
RyanTheTide
left a comment
There was a problem hiding this comment.
The base branch satchlj/port-307-20261104 (27d860d) is already in integration/aurora: it landed through #969 and #964 (integration a45bea1). Please retarget this PR to integration/aurora. Its only remaining commit is aef029a, and it merges cleanly there. CI will run once it targets integration.
Before it leaves draft, it needs a J615 result on 26.6.2 (25G83) firmware from that head:
- with
chosen.asahi,j615-25g83-experimental=1: the GPU starts on the Extended profile and GL/Vulkan run; - without the property:
loaded_profile()returns ENODEV, the driver does not bind, and the boot is otherwise unchanged.
Please also post the j615_requires_exact_optin_and_unambiguous_board and extended_admission_requires_exact_firmware_and_air results.
The 26.6.2 (25G83) IPSW ships the same GPU, DCP and PMP firmware for the 13-inch (J613) and 15-inch (J615) M3 Air, and the J615 ADT matches J613's in every node this path uses. Allow J615 onto the extended profile, but only with the same exact opt-in m1n1 and the DCP code already require (chosen asahi,j615-25g83-experimental = "1"), and only when the board identity is unambiguous. J613 behaviour is unchanged. This has not been qualified on J615 hardware. Signed-off-by: Satya Benson <satya@benson.earth>
c5a7262 to
7c6a1f7
Compare
|
@ericjuta @mmonari @M4Marvin @contursofran @omermuneer @AnotherCoolDude: you've each reported running Linux on a 15-inch M3 Air (J615). Would any of you be open to helping test this PR? It lets the J615 GPU start when Linux boots on the newer macOS 26.6.2 (25G83) firmware, which is what the 13-inch (J613) already uses. I only have a J613, so this can't merge without a result from a real J615. The catch is the firmware: from your reports you're all on the 14.7 OS firmware, so the first step would be moving your Linux install onto a 26.6.2 boot volume. That is more than booting a test kernel, and we're happy to help you through the whole process, step by step. Once on that firmware, the test itself is two boots of a kernel built from this branch:
If you're interested, reply here and we'll take it from there. No pressure if it's not a good time. |
|
Retargeted to integration/aurora; only the J615 opt-in change remains. Both |
|
I am down 100 percent, i am also dev, fyi just in case. |
@M4Marvin sweet thanks! It's me this time lol Sorry for the claudeslop above, Claude is supposed to identify itself when it posts on GitHub with my acct but apparently missed that I'll send you more info soon - are you on the Aurora discord? |
Not yet could you send me a link |
|
DjDeveloperr
left a comment
DjDeveloperr
left a comment
There was a problem hiding this comment.
Codex (GPT-6.1 Sol), reviewing on @DjDeveloperr’s behalf.
The exact opt-in, board-ambiguity and firmware gating look appropriately narrow, and host tests pass. Keep this draft until a real J615 on 25G83 demonstrates the exact opt-in Extended GL/Vulkan case plus absent/malformed-property ENODEV/no-bind controls. J613 boot/composition reports do not qualify this J615 path. I am coordinating volunteered hardware with ryan-orch-regress; no firmware upgrade or hardware run has been authorized by this review.
Stacked on #969 (port of omacom/linux-aurora#307); review that first.
Admit J615 only with an unambiguous board identity and the exact shared 26.6.2 experimental loader opt-in. Retain the newer #307 MMU correction. J615 hardware remains unqualified.
Port of omacom/linux-aurora#310. Depends on the port of omacom/linux-aurora#307; use the stack base above to review this layer. The original PR remains untouched.
Carried dependencies (original patch, complete message, authorship and dates preserved; 7.3 adaptations are separate commits):
e51f1082fa2384ef8e4d724dad52c9b317f6e45af192f61d8e864a57c40ee16c5bbae3f875c5823f60b3eae5bfa6b7d3d2666d8ca7a5c9dcc611f3f1e8023bba60e634b754f9d5cf8a31fe1856860486865371b461eccdbbee806b505f46bda16246daaee9d200311a5359dafc5391123e5ac297e5a4401080958cd5a4956cdb981acb5b18c64047548cba17f5af0ba04b7aa809ff54a8f092b08ba0@scottjones @chriskearney @DjDeveloperr @RyanTheTide
This carries Scott Jones's T8122 runtime and safety foundation from omacom/linux-aurora unaltered, because Aurora doesn't have it yet and this change builds on it. If you'd rather land it separately or differently, say so and I'll rework this.
This carries Chris Kearney's M3 native defaults and startup diagnostics from omacom/linux-aurora unaltered, because Aurora doesn't have it yet and this change builds on it. If you'd rather land it separately or differently, say so and I'll rework this.
This carries DjDeveloperr's M3 client parameter admission and timestamp adapter from omacom/linux-aurora unaltered, because Aurora doesn't have it yet and this change builds on it. If you'd rather land it separately or differently, say so and I'll rework this.
Aurora #783 held this foundation pending identity-selected constants and J613/J615 qualification. This port does not clear that hold.
Validation: arm64 defconfig + asahi.config, Image/modules/DTBs with BTF and KUnit enabled, LLVM 18/Rust 1.85.0, passed at stack top #310 (
151ddf9295cb), 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. M3 runtime, firmware, batching, power and paired UAPI fixtures passed; J615 hardware remains unqualified. No 7.3 hardware boot or deployment has been performed.