[pull] canary from vercel:canary#796
Merged
pull[bot] merged 8 commits intocode:canaryfrom Feb 17, 2026
Merged
Conversation
When a route has `unstable_instant` defined on any segment, setting
`prefetch={true}` on a Link to that route is now a no-op. The prefetch
behaves the same as the default auto strategy — cached content is
prefetched but dynamic content is deferred to navigation time.
A full static prefetch doesn't make sense for routes with instant
configs, since those segments use runtime prefetching. Rather than
silently doing something the developer didn't intend, we ignore the prop
entirely.
To detect this on the client, the server now sends a bitmask instead of
a boolean for the per-segment prefetch metadata. One bit tracks whether
the subtree contains any instant config, which propagates up so the
client can check it at the root without traversing.
This aims to avoid us having tests stall for upwards of 240s when a single assertion in the test stalls. This is very common currently. To also avoid wasting the entire time of resetting the test suite this also allows 1 retry for individual failed assertions. This won't always work since some assertions rely on previous setup so this also keeps the entire test suite retry handling as a fallback. This change is not applied for dev tests as they check HMR and compile lazily. ## Summary - keep elevated Jest timeout for test setup/startup in e2e-utils - enforce a 60s default timeout for individual `it`/`test` cases - apply one retry per individual test via `jest.retryTimes(1)` ## Verification - pnpm testonly test/development/gssp-notfound/index.test.ts - pnpm testonly test/development/enoent-during-require/index.test.ts - pnpm testonly test/e2e/custom-app-render/custom-app-render.test.ts Tested against our deploy tests to see if this helps: Current run [21967823508](https://github.com/vercel/next.js/actions/runs/21967823508) Jobs: 24 (with test-runner output: 20) Whole-suite retry starts: 11 (retry 1: 8, retry 2+: 3) Jobs that needed suite retries: 8 Unique test files retried: 7 Previous run [21964085768](https://github.com/vercel/next.js/actions/runs/21964085768) Jobs: 26 (with test-runner output: 20) Whole-suite retry starts: 18 (retry 1: 13, retry 2+: 5) Jobs that needed suite retries: 9 Unique test files retried: 12
## Summary Reworks Turbopack deferred entries in production build so deferred routes are processed last in the same project lifecycle, with `onBeforeDeferredEntries` executed between the non-deferred and deferred phases. ## What changed - Moved deferred build phasing into Rust in `project_write_all_entrypoints_to_disk`: - non-deferred phase - `onBeforeDeferredEntries` - deferred phase - Kept callback invocation on the Rust side via the existing node-worker callback plumbing. - Narrowed invalidation after `onBeforeDeferredEntries` to deferred app source subtrees instead of invalidating the whole project filesystem cache. - Added `DiskFileSystem::invalidate_path_and_children_with_reason(...)` in `turbo-tasks-fs`. - Derived deferred invalidation dirs from deferred routes’ `original_name` values. - Retained a full-invalidation fallback only when app-dir path resolution is unavailable. - Deferred route selection still includes both app pages and app route handlers. ## Validation - `cargo check -p next-napi-bindings -p turbo-tasks-fs` - `pnpm test-start-turbo test/e2e/deferred-entries/deferred-entries.test.ts` (21/21 passing) - Validated against [workflows repo](vercel/workflow#1058) and v0
This ensures we always set this env variable so that it overrides any flag opt ins when we specifically want to test with or without the adapter.
Stacked on #89971 This controller was introduced to help with dev time validation. It models stage transitions in a helpful way that will become more useful soon as we get more advanced with what kinds of sync IO are considered errors. To prepare for this we will update the implementation to use the stage controller for build and runtime prerendering not just dev We don't yet use it for static prerendering because that one is so simple and there is really no stage beyond the first one and then an abort so we will keep it as is Additionally, the private cache store no longer has the `runtimeStagePromise`. when you first enter this store type during a runtime prerender we wait until the outer store's runtime stage has begun. Previously the proimse signifying this stage transition was included in the private cache store itself so nested caches could similarly wait, however you can't actually get to a nested cache until after it is resolved b/c the outermost one blocks.
Stacked on #89972 The pipelining function has some awkward structure. This change cleans up the structure a bit Notably there is a potentially meaningful change here where the final user provided task in runInSequentialTasks gets the same Fast Immediate treatment as the other tasks do. There is an implicit final task which does the assertion that there are no remaining immediates unflushed and then resolves the original tasks value.
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 subscribe to this conversation on GitHub.
Already have an account?
Sign in.
4 participants
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.
See Commits and Changes for more details.
Created by
pull[bot] (v2.0.0-alpha.4)
Can you help keep this open source service alive? 💖 Please sponsor : )