JIT: hot-reload non-leaf files via single-module incremental split#194
Merged
Conversation
bitmapImageRepForCachingDisplay inherits the host window's backingScaleFactor, so on a Retina (2x) display the snapshot came out 800x1200 instead of the logical 400x600, making output depend on which machine the daemon ran on. Build the NSBitmapImageRep manually at the point bounds so capture is 1x and reproducible across machines. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The structural split compiled the bulk into a prebuilt stable module that the hot file imported. That is one-directional, so a non-leaf hot file (one the bulk references, e.g. a previewed view used by another preview) failed with "cannot find <type> in scope" when the bulk compiled without it. Detect the non-leaf case (the stable-module compile fails) and compile the whole target module incrementally instead: the editable overlay plus the other sources under one module name, where the Swift driver recompiles only the hot file and reuses the bulk objects. References resolve in both directions. Because the overlay then uses the target's own module name, its @observable DesignTimeStore would re-register across generations, so non-leaf builds respawn the agent each structural edit. hotReloadStructural now exercises this on the JIT daemon. Its sync marker moves from "Compiled:" (only the non-JIT recompile logs it) to "Reloaded", which both daemon kinds log on a successful structural edit. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
lookupInitialized read session->initialized outside the lock with no re-check, so two threads could both pass the guard and double-initialize the JITDylib (and the bool itself was a data race). Hold the lock across the check and the init so the flag is only touched under the lock. PreviewsAnonymousMapper::prepare and ::initialize decremented the upper_bound iterator with no bounds check, which is undefined behavior if the reservation map is empty or the address precedes all reservations. Guard r == begin(): prepare returns nullptr, initialize reports an Expected error. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
… plan Adds the "Update 2026-06-09" section that supersedes the stale "Immediate next step" pointer and the CI-green claims: PR #190 merged, GitHub Actions CI is disabled, and the merge bar is now local (unit tests plus the integration-test skill). Documents the non-leaf duplicate/missing-symbol bug, the incremental whole-module fix this branch lands, its tradeoffs, and the verification table. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
ImageRenderer cannot rasterize AppKit-backed SwiftUI: with
NavigationStack { List } as the root it returns a nil cgImage (the
agent entry fails with status -1) and with any modifier wrapping the
List it returns a tiny blank raster at ideal size. Every real preview
hit one of the two; the daemon then either reported "Recompilation
failed" or served a blank snapshot. The cases slipped through because
the existing assertions only checked "snapshot changed" / "PNG
non-empty" and the session-start render goes through the dylib window
path, so the agent had never rendered an unedited body.
renderPreviewToFile now hosts the view in a borderless NSWindow
positioned off-screen via NSHostingView, which gives AppKit-backed
views a real backing hierarchy, and captures at a deterministic 1x
into an NSBitmapImageRep the same way Snapshot.capture does. No
run-loop settle is needed; cacheDisplay draws synchronously after
layoutSubtreeIfNeeded.
nonLeafRendersUneditedPreviewInAgent guards the regression: it renders
the unedited non-leaf ToDoView.swift (NavigationStack + List) through
compileObjectForJIT and the reloader, exactly the path a cross-file
bulk edit takes as a session's first structural reload.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
snapshot() prefers the agent-rendered PNG whenever agentImagePaths has an entry, but only the JIT reload paths wrote it and nothing cleared it when the dylib window path re-rendered. After a session became agent-backed, preview_switch and preview_configure updated the window while snapshots kept serving the old agent PNG. loadPreview now removes the entry, making the window the source of truth again until the next JIT reload repopulates it. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
"/third_party/" only matches directories. Worktrees point third_party at the main checkout's prebuilt LLVM via a symlink, which therefore showed as untracked. Dropping the trailing slash covers both. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Problem
main(a5b58cc, the #190 merge) shipped the JIT split-compile path with onlythe leaf case handled. The split prebuilds a stable module from the target's
other sources and
@testable imports it from the edited file. When the editedfile is non-leaf (another file in the target references it, e.g.
ToDoView.swiftis used byToDoProviderPreview.swift), the stable compileexcludes the hot file but the bulk still references it, so it fails with
cannot find 'ToDoView' in scopeand the hot reload times out.Running the full local merge bar (the
integration-testskill) against thisbranch then exposed two further agent-render bugs that
mainalso has:ImageRenderercannot rasterize AppKit-backed SwiftUI. WithNavigationStack { List }as the root it returns a nilcgImage(the agententry fails with status -1, surfaced as
Recompilation failed: JIT render entry returned non-zero status -1on any cross-file edit that was asession's first structural reload). With a modifier wrapping the
Listitreturns a tiny blank raster instead. Existing tests passed because they only
asserted "snapshot changed" / "PNG non-empty".
snapshot()prefers theagent-rendered PNG once a session is agent-backed, but the dylib window
paths (
preview_switch,preview_configure) never cleared it, sosnapshots kept serving the pre-switch image.
Fix
Detect the non-leaf case (the stable-module compile throws) and compile the
whole target module incrementally instead: the edited file as an overlay
plus the other sources under one module name, where the Swift driver recompiles
only the hot file and reuses the bulk objects. References resolve both
directions, so the hot file lives in exactly one module.
Because the overlay then uses the target's own module name, its
@ObservableDesignTimeStorewould re-register across generations, so non-leaf buildsrespawn the JIT agent each structural edit (
requiresFreshAgent).For the render bugs:
renderPreviewToFilenow hosts the view in a borderless off-screenNSWindowviaNSHostingViewand captures at a deterministic 1x, the samemechanism as
Snapshot.capture. AppKit-backed views get a real backinghierarchy and render with full content.
loadPreviewdrops the session'sagentImagePathsentry, making the windowthe source of truth again until the next JIT reload repopulates it.
Design notes / known tradeoffs
still serves the common case (editing the preview-bearing leaf file).
(
bulkIsNonLeaf). A real error in another bulk file would pin the session tothe slower whole-module path. Acceptable: it degrades speed, not correctness,
and the error still surfaces. Follow-up candidate.
Verification
hotReloadStructural(editsToDoView.swift, the non-leaf file) passesagainst the JIT daemon, where
maintimes out at 90s.nonLeafRendersUneditedPreviewInAgentrenders the unedited non-leafToDoView.swiftthroughcompileObjectForJIT+ the reloader (the exactcross-file-first path), failing on
main'sImageRendererentry.PreviewsJITLinkTests: 51/51 serially. Touched-module unit suites(Core/MacOS/Engine): 320/320. macOS CLI command suites: 46/46 serially.
macOS MCP integration suite: 7/7.
integration-testskill, macOS scope, all four examples (SPM, Bazel,xcodeproj, xcworkspace): rendering, literal + cross-file hot reload,
multi-preview switch with invalid-index rollback, trait injection +
persistence across switch, PreviewProvider. Daemon log clean of reload
failures.
Commits
Snapshot: pin window capture to deterministic 1x(independent, also strandedoff
main)JIT: hot-reload non-leaf files via single-module incremental split(the fix)JIT C++: fix lookupInitialized lock and anon-mapper bounds(JIT-linkhardening)
Docs: record non-leaf fix, CI removal, and local merge bar in Phase 3 planJIT: render agent preview in an off-screen NSWindow, not ImageRendererHostApp: drop stale agent PNG when the window reloads a dylibgitignore: match the third_party symlink in worktrees🤖 Generated with Claude Code