Skip to content

cmux-next: pin CCmuxAppFFI to cmux-app-ffi-b24aac2c507d072ac3b5cf35fb3c3b93bd039acf - #35853

Merged
lawrencecchen merged 1 commit into
feat-cmux-nextfrom
app-ffi-repin
Nov 4, 2026
Merged

lawrencecchen merged 1 commit into
feat-cmux-nextfrom
app-ffi-repin

Conversation

@manaflow-cmux-ffi-release

@manaflow-cmux-ffi-release manaflow-cmux-ffi-release Bot commented Nov 4, 2026 •

Copy link
Copy Markdown

Pins CCmuxAppFFI to cmux-app-ffi-b24aac2c507d072ac3b5cf35fb3c3b93bd039acf (checksum 5aa92a5b76b173dce2a254ab9e437c0b086d3b4d07fd9754ab3d72b0f14b9481), published by run https://github.com/manaflow-ai/cmux/actions/runs/41899051904, which checked it with check-app-ffi-pin.sh --verify-release. A newer publish updates this branch.

Changelog

none


Autofix with [code]smith
Open in [code]smith · Need help on this PR? Tag @codesmith-bot with what you need.


Summary by cubic

Updates the pinned CCmuxAppFFI binary target in CmuxNext/Package.swift to the newer cmux-app-ffi-b24aac2c507d072ac3b5cf35fb3c3b93bd039acf release, refreshing the download URL and checksum. No changelog entries; the newer publish supersedes this pin.

Written for commit acccfaa. Summary will update on new commits.

View guided diff Turn on auto-fix

@github-actions

github-actions Bot commented Nov 4, 2026 •

Copy link
Copy Markdown
Contributor

cmux status (one comment, edited in place; each line is updated by the job that owns it)

Review acccfaa: HQ feed card · gallery before/after runs · checks

App of this head (app.zip, kept 1 day)

gh workflow run reload-build.yml -R manaflow-ai/cmux --ref feat-cmux-next -f tag=pr35853 -f ref=app-ffi-repin -f platform=macos -f backend_mode=local

Shipped in:

@github-actions github-actions Bot added the risk:normal Risk tier (docs/ci/risk-tiers.md) label Nov 4, 2026
@lawrencecchen

Copy link
Copy Markdown
Contributor

Review (independent review subagent): SHIP at acccfaa

What I checked (read-only; nothing was built or run locally):

Checks on this head:

  • "cmux-next checks" is green, including "Check the app FFI pin" ("sources match b24aac2", "release built from ... with checksum 5aa92a5b..."). On the base, the same pin check reports drift from the old pin df4352d75349 (push runs for 036f8aa53845 and b24aac2c507d), so this PR fixes that base red.
  • Compile checks pass: "cmux-next app compile (build fleet)", "cmux-next fleet Release compile", "cmux-next/compile", "fleet/test-compile (advisory)". "risk evidence", "fast gates" and "cmux-next CI guard group" pass too.
  • No red check. Still pending when I posted this: "cmux-next swift test", "cmux-next daemon tests", "cmux-next fast tier", "nightly-guards / workflow-guard-tests / release-notary". I did not wait for them. The nightly-guards check was a base finding on cmux-next: pin CCmuxAppFFI to cmux-app-ffi-df4352d75349a4374f29caac31458c83d4a0c6dc #35420 and this PR changes no workflow.

@lawrencecchen lawrencecchen added the ready-to-land Reviewed at this head; next-autoland merges it into feat-cmux-next when its only reds are base reds label Nov 4, 2026
@lawrencecchen
lawrencecchen merged commit 329d06e into feat-cmux-next Nov 4, 2026
229 of 244 checks passed
@lawrencecchen
lawrencecchen deleted the app-ffi-repin branch November 4, 2026 20:40
lawrencecchen added a commit that referenced this pull request Nov 4, 2026
…he app's one input router (cmux_input_*) (#35875)

* cmux-next: pin CCmuxAppFFI to cmux-app-ffi-b24aac2c507d072ac3b5cf35fb3c3b93bd039acf

* R-INPUT S2b step B: every window's focus machine keeps its focus in the app's one input router (cmux_input_*)

Step B of the FFI two-step (step A #35641, re-pin #35853): FocusMachine calls the input handle
(`cmux_input_focus_*`, one window key per machine) instead of a per-window `cmux_focus_*` handle.
The router is made once at the composition root (AppServices.focusInput, FocusInput in
CmuxNextFocus) and given to each window's FocusCoordinator through WindowState. Pure refactor:
the same requests and the same reply bytes (step A's corpus test), so every focus rule is
unchanged.

- FocusInput owns the CmuxInput handle (memory: the Mac's physical memory, which bounds the key
  lanes). Each machine opens a window under a key no other machine of that router had, and closes
  it when freed, so a window id the app reuses (a restored closed window) never meets the state
  of the machine it replaces, whichever is freed first.
- A machine made with no input (tests, the journal replay) gets a router of its own.
- FocusMachine's Close Tab aim calls move to FocusMachine+CloseAim.swift (A12 follow-up); the
  file shrinks from 467 to 458 lines.
- FocusResponderClassifier.describe moves to the package as FocusEvent.Responder.debugName
  (CmuxNextApp: -10 lines).

Tests: FocusInputSharedRouterTests: a machine in the shared router reduces exactly as a machine
alone (state and effects after every event) while another window acts between its events; a new
machine never meets a freed machine's focus; every request goes through the input handle (the
close-focus policy, the Close Tab aim, the history, the remembered panes, the checkpoint; a
checkpoint that does not decode is refused and leaves the state as it was: review notes of
#35641).

* FocusInput: the memory default is read in the init (a default argument may not name an internal import)

* test(focus): the input-handle request test follows PaneCloseAim's documented order (the focus moves, then the frame shows) and compares each answer with a machine alone

---------

Co-authored-by: manaflow-cmux-ffi-release[bot] <339941766+manaflow-cmux-ffi-release[bot]@users.noreply.github.com>

This branch was previously deployed

1 inactive deployment
ci-routing-agent — acccfaac Deployed Nov 4, 2026 by manaflow-cmux-ffi-release[bot] via wait for same-key tree publisher #112943
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ready-to-land Reviewed at this head; next-autoland merges it into feat-cmux-next when its only reds are base reds risk:normal Risk tier (docs/ci/risk-tiers.md)

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant