Skip to content

Add MacBookPro13,1 hardware support - #7732

Draft
gjinali wants to merge 5 commits into
omacom:quattrofrom
gjinali:codex/macbookpro13-1-distribution-support
Draft

gjinali wants to merge 5 commits into
omacom:quattrofrom
gjinali:codex/macbookpro13-1-distribution-support

Conversation

@gjinali

@gjinali gjinali commented Aug 22, 2026 •

Copy link
Copy Markdown

Summary

  • install model-gated Cirrus audio and FaceTime HD camera packages during fresh setup and hardware replay
  • migrate existing MacBookPro13,1 systems away from the obsolete SPI DKMS package while retaining in-tree applespi
  • include the offline compatibility packages and document the MacBookPro13,1 normal-boot support boundary

Depends on omacom/omarchy-pkgs#182.

Validation

  • MacBookPro13,1 focused shell test: 17/17 passed
  • package and offline DKMS validation passed on Linux 7.1.8
  • live migration, reboot, audio, microphone, camera, keyboard, trackpad, and keyboard-backlight checks passed
  • official hardware replay passed with before/after state unchanged except timestamp

Deep suspend and hibernation are not supported on MacBookPro13,1. The reference-device deep-suspend test left both analog audio outputs silent until reboot, so no experimental downstream workaround is included. See the upstream deep-suspend audio proposal and jack-detection issue for technical background.

Copilot AI balanced review requested due to automatic review settings August 22, 2026 02:44

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

Adds targeted MacBookPro13,1 support while preserving existing behavior for other Apple models.

Changes:

  • Adds model-gated audio and camera installation.
  • Migrates existing systems from the obsolete SPI DKMS driver.
  • Adds offline packages, documentation, and focused tests.

Tip

If you aren't ready for review, convert to a draft PR.
Click "Convert to draft" or run gh pr ready --undo.
Click "Ready for review" or run gh pr ready to reengage.

Reviewed changes

Copilot reviewed 8 out of 8 changed files in this pull request and generated no comments.

Show a summary per file
File Description
install/hardware/all.sh Runs the new hardware installers.
install/hardware/apple/fix-spi-keyboard.sh Uses in-tree applespi for this model.
install/hardware/apple/install-macbookpro13-1-audio.sh Installs Cirrus audio support.
install/hardware/apple/install-macbookpro13-1-camera.sh Installs camera driver and firmware.
install/omarchy-other.packages Caches compatibility packages offline.
migrations/1787257711.sh Migrates existing installations.
manual/44-mac-support.md Documents support and limitations.
test/shell.d/macbookpro13-1-hardware-test.sh Tests setup, migration, and documentation contracts.

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

@omarchybot

Copy link
Copy Markdown
Collaborator

Reviewed at 58bad8242bd38bdb52c5dc3c836730add5b7a5d9 against quattro. The tests pass, but this can't land yet: its packages don't exist, and three things below need changing. Separately, the maintainer needs to decide how it relates to other open pull requests that cover the same hardware.

What ran. On a disposable Omarchy worker, ./test/cli, test/shell.d/macbookpro13-1-hardware-test.sh (17/17) and the related brcmfmac-supplicant, migrate-scope, migrate-wrapper, systemd and t2-hardware tests all passed. The hardware behaviour itself (audio, camera, keyboard on a real MacBookPro13,1) can't be reproduced on a worker, so nothing here confirms it. Your own testing on the A1708 is described in the pull request body. Under the project's rules, confirmation for a hardware fix has to come from a maintainer or the person who reported the problem, not the pull request's author.

Found:

  • The packages aren't published yet. snd-hda-macbookpro-dkms, facetimehd-dkms and facetimehd-firmware aren't in the edge, rc or stable Omarchy repositories (checked against each channel's omarchy.db), and Add MacBookPro13,1 camera and audio packages omarchy-pkgs#182 is still open. If this merged first, install/omarchy-other.packages:69 would break the ISO build for every machine, not just Macs. On a MacBookPro13,1, hardware setup and migrations/1787257711.sh:6 would fail. A migration that fails blocks every migration after it. So use uwsm to properly wrap applications as systemd units #182 has to merge and publish before this one.
  • The migration's timestamp is older than quattro's newest migration (1787257711, Aug 20, against 1790542069). agents/skills/migrations.md says migrations are strictly ordered, so it needs a current timestamp from omarchy-dev-add-migration when it lands.
  • migrations/1787257711.sh:3 reads its marker path from OMARCHY_MACBOOKPRO13_1_MIGRATION_MARKER. Point it at a file that already exists and the migration skips all its work but still records itself as complete. Otherwise it chooses where a root-owned file gets written. That override exists only for the test and shouldn't be in the shipped migration. Checking state instead of a marker would also keep the second user on a machine from repeating the work: for example, skip when the packages are already present and macbook12-spi-driver-dkms is gone.
  • The test can't catch a wrong model check. The omarchy-hw-match stub at test/shell.d/macbookpro13-1-hardware-test.sh:23 ignores its pattern and returns whatever MATCHING_HARDWARE says. On the worker I changed the check in both installers and the migration from '^MacBookPro13,1$' to '.', which matches every machine, and all 17 tests still passed. For the "other models untouched" tests to mean anything, the stub should check the pattern against a fake product name.

The manual's corrected T1 list is right: A1708 is the two-port MacBookPro13,1 without a T1, A1706 the four-port 13-inch, A1707 the 15-inch.

Overlapping pull requests. These cover the same hardware in different ways and would conflict or duplicate work if both landed:

#12968 (draft, audio recovery on this model) complements this one rather than competing with it. Which approach to take is the maintainer's call, so none of these was brought to ready.

Second opinion. Codex Medium reviewed it too. It agreed on the package ordering, the migration timestamp and the manual (independence isn't guaranteed). It raised the marker override and the test stub ignoring its pattern on its own, and I confirmed both against the code and on the worker. It couldn't read the competing diffs, so it didn't rank them.

Next: waiting on the maintainer to choose between the overlapping approaches and to land omacom/omarchy-pkgs#182. The three fixes above are yours, if this is the approach they pick.

@greptile-apps

greptile-apps Bot commented Oct 18, 2026 •

Copy link
Copy Markdown
Contributor

RetriggerView in GreptileConfidence Score: 5/5

[Medium impact] No actionable issue remains from the changes since the previous review.

Summary

Adds MacBookPro13,1 support for normal boot.

  • MacBookPro13,1 setup adds normal-boot audio and camera support.
  • The MacBookPro13,1 migration adds support and switches the keyboard driver.

No new actionable findings were accepted. No previous review threads were supplied.

Reviews (2) · Last reviewed commit: "Move the kernel-module update gate to it..." · Reviewed by Greptile

@omarchybot omarchybot removed compatibility Make Omarchy work better on everything! bug Something isn't working labels Oct 18, 2026
@omarchybot

Copy link
Copy Markdown
Collaborator

Looked again at e0ad4dc645eecad80441ce5b46980047cc0e6d0f, the new commit "Block reboot when required drivers are missing". This pass only read the change. Nothing ran on a worker, because the classification below isn't settled.

Classification is unsettled, so the earlier bug and compatibility labels have come off. I read it as a hardware compatibility fix. Codex Medium, asked separately without my answer, called it an enhancement. Both readings hold up, because the new commit changes omarchy update for every machine: bin/omarchy-update-verify-kernel-modules is a new step in the update sequence, and bin/omarchy-update-restart now refuses to offer a reboot when it fails. It does nothing unless a package-owned verifier is installed under /usr/lib/omarchy/, and only omacom/omarchy-pkgs#182 ships those. Even so, it's a change to the update flow in its own right. A pull request that both fixes and adds is two pull requests under the project's rules, so the maintainer may want the update gate split out.

Still open from the earlier review, unchanged at this head:

  • Add MacBookPro13,1 camera and audio packages omarchy-pkgs#182 is still open, so snd-hda-macbookpro-dkms, facetimehd-dkms and facetimehd-firmware aren't published yet, and install/omarchy-other.packages would break the ISO build if this landed first.
  • migrations/1787257711.sh still has a timestamp older than quattro's newest migration (1792334754).
  • migrations/1787257711.sh:3 still takes its marker path from OMARCHY_MACBOOKPRO13_1_MIGRATION_MARKER.
  • The omarchy-hw-match stub at test/shell.d/macbookpro13-1-hardware-test.sh:23 still ignores its pattern, so a wrong model check would pass.

Also new: this branch no longer merges cleanly into quattro. The conflict is in docs/update-process.md.

On the new commit: the interface it expects (--modules-root plus targets on stdin, installed as /usr/lib/omarchy/verify-snd-hda-macbookpro and verify-facetimehd) matches the verifiers in omacom/omarchy-pkgs#182 at 373ffaa. Reading it turned up no defect.

Next: waiting on the maintainer to settle what this pull request is, and whether the update gate belongs in it. It also still needs them to choose among the overlapping Intel Mac pull requests (#12684, #12686, #12689, all drafts) and to land omacom/omarchy-pkgs#182.

The gate now lives in omacom#15905, rebuilt on quattro's boot
check and extended to the linux-omarchy kernel.

This reverts commit e0ad4dc.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
@gjinali

gjinali commented Oct 18, 2026

Copy link
Copy Markdown
Author

Thanks for both reviews. Here's where this stands now.

Update gate split out. The kernel-module check now has its own PR: #15905. It is rebuilt on current quattro's boot-check gate (omarchy-update-boot / boot_verified) rather than adding a separate step, so the separate omarchy-update-restart refusal is gone. It checks the stock linux-omarchy kernel and the Arch linux kept as its fallback; the version reviewed here only checked linux. That removes the update-flow change, and the docs/update-process.md conflict, from this PR.

Packages first. omacom/omarchy-pkgs#182 has been updated for linux-omarchy. It also moves facetimehd-dkms to upstream 0.7.2, which builds on Linux 7.2 without our patch. This PR stays blocked until #182 is merged and published, as you said.

Findings accepted. I'm not disputing any of the three code findings. If this is the approach you pick, I'll:

  • regenerate the migration with omarchy-dev-add-migration when it lands;
  • replace the marker override with a state check: skip when both packages are present and macbook12-spi-driver-dkms is gone;
  • make the omarchy-hw-match stub match its pattern against a fake product name, so that '.' fails the "other models untouched" tests.

Overlap with the Intel Mac series. That choice is yours. I've offered my MacBookPro13,1 (A1708) for attended testing of #12686, #12689 and #12684, and proposed #182 as the package layer for this model on the series' package PRs. Until you decide, I'm converting this to a draft so it doesn't read as ready.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants