Skip to content

Document T1 Touch Bar t1bridge restore on Mac support page - #16101

Open
Alkarth2010 wants to merge 5 commits into
omacom:quattrofrom
Alkarth2010:docs/t1-touchbar-s2idle
Open

Alkarth2010 wants to merge 5 commits into
omacom:quattrofrom
Alkarth2010:docs/t1-touchbar-s2idle

Conversation

@Alkarth2010

@Alkarth2010 Alkarth2010 commented Oct 20, 2026 •

Copy link
Copy Markdown

The Mac support page currently says the T1 Touch Bar is non-functional. On MacBookPro13,2 (2016 13-inch with Touch Bar) running Omarchy 4.0.6 / kernel 7.2.8-9-omarchy, the bar, webcam, and function-row keys work after restoring this Mac's Apple EFI data and installing t1bridge.

This PR documents that workaround on the T1 known-issues list, in the same shape as the CS8409 speaker write-up:

  • Omarchy's built-in Touch Bar stack is T2-only; there is no physical Esc/F-row on these chassis
  • If USB is 05ac:1281 or EFI/APPLE/EMBEDDEDOS is missing, restore with t1-revive first (irreversible without an EFI backup; no --no-confirm)
  • Add the signed t1bridge repo, omarchy update, then pacman -S t1bridge t1bridge-dkms t1bridge-omarchy with linux-omarchy-headers (omit fingerprint packages unless Touch ID is wanted)
  • Enable t1-touchbar-hw, reboot for group t1bridge, enable the user renderer
  • After suspend, taps can work while the strip stays dark; hidraw report 3 unpark + s2idle workaround at https://github.com/Alkarth2010/t1-touchbar-s2idle
  • omarchy-hw-external-monitors treating card*-USB-* as a dock

This does not ship t1bridge in the ISO. Same T1 hardware exists on other 2016–2017 Touch Bar Macs; the write-up only claims MacBookPro13,2. Sound is left as-is so it does not collide with #15776.

t1bridge: https://github.com/standardagents/t1bridge
t1-revive: https://github.com/niconistal/t1-revive
Sleep/wake issue: standardagents/t1bridge#18

Alkarth2010 added 2 commits October 21, 2026 01:28
The T1 bar is not dead on Omarchy: t1bridge lights it after EmbeddedOS is present, and hidraw report 3 unparks a dark strip after s2idle. Document that on the T1 known-issues list for MacBookPro13,2.
Cover recovery-mode 05ac:1281 / missing EMBEDDEDOS, t1-revive, the signed
t1bridge packages on Omarchy, group membership, the user renderer, and
the s2idle hidraw unpark. Only claimed on MacBookPro13,2.
@Alkarth2010 Alkarth2010 changed the title Document T1 Touch Bar t1bridge and s2idle unpark on Mac support page Document T1 Touch Bar t1bridge restore on Mac support page Oct 20, 2026
@greptile-apps

greptile-apps Bot commented Oct 20, 2026 •

Copy link
Copy Markdown
Contributor

RetriggerView in GreptileConfidence Score: 5/5

[Medium impact] This documentation change appears safe to merge.

Summary

The Mac support page documents a community T1 Touch Bar workaround and adds safer recovery guidance.

  • T1 Mac owners get steps to preserve and recover Apple EFI data.

No new actionable issues found. The previous incorrect tiny-dfr claim has been removed.

Reviews (3) · Last reviewed commit: "Clarify T1 firmware recovery and optiona..." · Reviewed by Greptile

Comment thread manual/44-mac-support.md Resolved
Co-authored-by: greptile-apps[bot] <165735046+greptile-apps[bot]@users.noreply.github.com>
@omarchybot omarchybot added the documentation Improvements or additions to documentation label Oct 22, 2026
@omarchybot

Copy link
Copy Markdown
Collaborator

Reviewed by Claude Opus 5.5 and, as a second opinion, Codex Medium (Codex agreed with the findings below and added the DKMS and t1-revive points; its independence from my reasoning is not guaranteed). This is documentation, so nothing was run: I read the change against the repository at 493c2d53 and against the t1bridge and t1-revive READMEs. Neither of us could check the hardware behaviour itself.

What checks out against the code

  • linux-omarchy-headers is the right headers package: a T1 Mac runs linux-omarchy, not linux-t2.
  • omarchy update before a plain pacman -S --needed is what t1bridge's own Omarchy note says to do.
  • The sysusers, tmpfiles, group, service and status commands match t1bridge's README, and so does the signing fingerprint.
  • The fingerprint warning is accurate: bin/omarchy-setup-security-fingerprint installs libfprint-git fprintd and accepts package conflicts, so it would replace t1bridge's matched pair.
  • The external-monitor point is accurate: bin/omarchy-hw-external-monitors only skips eDP, LVDS and DSI connectors, so a connected card*-USB-* Touch Bar panel counts as an external monitor and the lid-close lock is skipped. That is a bug in Omarchy rather than something to document around. T2 Touch Bars use the same appletbdrm driver, so they are probably affected too, though that has not been checked on hardware.

What needs a decision before this can go in the manual

  • The page would send users to a third-party pacman repository, a firmware-restore tool (t1-revive, about a month old) whose effects can't be undone without a backup, and a root sleep hook in Alkarth2010/t1-touchbar-s2idle, a repository created the same day as this PR. Recommending any of these in the official manual is the maintainer's call, not something a review can settle.
  • The recovery advice doesn't match upstream. When the Apple EFI data is missing, t1bridge's README says to restore it by doing a full macOS first boot, then to back up what it regenerates. This PR sends readers to t1-revive instead and doesn't say it is departing from upstream. It also treats 05ac:1281, or a missing /boot/EFI/APPLE/EMBEDDEDOS, as proof the data is gone, before checking for a backup. If the t1-revive route stays, it should carry t1-revive's own precautions (stay at the machine, mains power, preflight, an off-disk backup), not only "do not pass --no-confirm".

For the author, whichever way that goes

  • The install sequence leaves out upstream's step of confirming that DKMS and the initramfs/UKI build succeeded before rebooting.
  • The suspend bullet reads as debugging notes: HID report bytes, the usbfs EBUSY cause, systemd's hook directory. The manual is end-user documentation, and t1bridge itself lists T1 suspend/resume as not working. A short line saying suspend does not bring the bar back, with a link to t1bridge#18, would fit the page better than the mechanism and an unconditional "use s2idle".
  • In Markdown, the paragraphs and code blocks after the first bullet are not indented, so they end the list and render outside it, and the next two bullets start a new list. A ##### Touch Bar subsection under Known Issues would keep the list intact.
  • This conflicts with Document T1 CS8409 speaker workaround on Mac support page #15776, which edits the same list. Whichever lands second will need a rebase.

Nothing was pushed to this branch, because the remaining fixes depend on the decision above. This is now waiting on the maintainer to decide whether the manual should recommend these third-party sources.

omarchybot and others added 2 commits November 18, 2026 01:33
Put the same-Mac EFI backup requirement before the installer erases the disk. A missing directory at one mount point does not establish data loss, so check preserved EFI partitions and backups before firmware recovery. Separate optional Touch ID setup from the Touch Bar procedure and require successful DKMS and boot-image builds before rebooting.

Keep the sleep workaround limited to the reported configuration and keep implementation details out of the user manual.

Co-Authored-By: Maaz Chowdhry <maazchowdhry2001@gmail.com>
Co-Authored-By: Codex High <noreply@openai.com>
Co-Authored-By: Opus 5.5 High <noreply@anthropic.com>
Co-Authored-By: muse-wonder via pi, thinking high <noreply@openrouter.ai>
The Touch ID importer reads calibration data rather than restoring the EFI boot tree. Say so explicitly and direct Touch ID readers through the full upstream installation procedure, including the earlier package and socket prerequisites.

Describe community reports without assuming pull request context. Link the recovery source review, document session startup and repository preservation, and identify the sleep workaround's broader power changes.

Co-Authored-By: Codex High <noreply@openai.com>
Co-Authored-By: Opus 5.5 High <noreply@anthropic.com>
Co-Authored-By: muse-wonder via pi, thinking high <noreply@openrouter.ai>
@omarchybot omarchybot added the verified Omarchy Triage has verified that this issue is ready for final review label Nov 18, 2026
@omarchybot

Copy link
Copy Markdown
Collaborator

Pushed documentation corrections through ba4e7a6. The pre-install EFI backup warning from #16673 is now included, with co-authorship credit to @heymaaz. That PR's dual-boot and T2 material remains separate.

The page now distinguishes restoring EFI boot files from importing Touch ID calibration, checks preserved EFI partitions and same-Mac backups before firmware regeneration, and sends Touch ID readers through the complete upstream setup. It also requires successful DKMS and boot-image builds before rebooting, keeps the suspend report specific to its tested configuration, explains the workaround's wider power changes, and fixes the Markdown list structure.

Codex High checked the documentation against the repository, T1Bridge's installation/setup and recovery guides, t1-revive's precautions, and the sleep/lid workaround's README and helpers. The package repository metadata lists the documented packages, including t1bridge-omarchy 0.2.2-1. Opus 5.5 High and muse-wonder via pi, thinking high reviewed the final commit and found no further factual corrections to make; their independence from this review is not guaranteed. Only Markdown changed, so this was checked by reading and git diff --check; no worker, suite or hardware test ran. The verified label records those documentation checks.

This still waits on the maintainer to decide whether the official manual should recommend the third-party package repository/signing key, t1-revive firmware recovery, and the community sleep/lid workaround. The documentation corrections do not settle that support decision. T1Bridge still lists system suspend/resume as unsupported.

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

Labels

documentation Improvements or additions to documentation verified Omarchy Triage has verified that this issue is ready for final review

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants