- 24 Jan, 2025 33 commits
-
-
vercel-release-bot authored
-
Tobias Koppers authored
### What? * more memory efficient AggregationUpdateQueue by using SmallVec and boxing some job types * avoid allocation in CaptureFuture by avoiding the Arc and the Mutex and using a thread local reference
-
Benjamin Woodruff authored
Allows tasks to be marked as local using `#[turbo_tasks::function(local)]`. Local tasks are cached only for the lifetime of the nearest non-Local parent caller. Local tasks do not have a unique task id, and are not shared with the backend. Instead they use the parent task's id and store data in the backend-agnostic manager. This is useful for functions that have a low cache hit rate. Those functions could be converted to non-task functions, but that would break their function signature. This provides a mechanism for skipping caching without changing the function signature. Local tasks are already used for argument resolution, this allows other arbitrary functions to be opted-in.
-
JJ Kasper authored
This improves the build script quite a bit and makes dev editing much faster. Also fixed excludes causing webpack/babel compiled packages to be bundled into the runtime un-necessarily reducing from `10MB` -> `1.2MB`. Before: `Finished next_bundle in 9.67s` After: `Finished next_bundle in 1.27s` Validated here https://github.com/vercel/vercel/actions/runs/12957911052/job/36147343396?pr=12914
-
Tobias Koppers authored
### What? * move code from macro into generic struct * move dynamic storage access into separate struct * generate inline storage for common data
-
Tobias Koppers authored
### What? This reduces the overallocated memory by InnerStorage instances in the DashMap capacity
-
Benjamin Woodruff authored
A function returning an `OperationVc` can call a transient function or have a transient argument. IMO, `TaskInput` should not have default method implementations to avoid potential footguns like this.
-
Benjamin Woodruff authored
We only read data owned by the the currently-executing task when reading from a local Vc, so: - A separate "untracked" API doesn't make sense. There's never any tracking needed for reading local outputs or cells. (would we mark the current task as dependent on itself?) - Allowing a `consistency` argument isn't useful because you can't strongly consistently await your own task. - We don't need to call `notify_scheduled_tasks`, we're not adding any new task dependencies.
-
Benjamin Woodruff authored
refactor(turbo-tasks): Merge dynamic_call/native_call with dynamic_this_call/native_this_call (#75253) The `_this` argument variants of these functions are 99% of the same, so this deduplicates them.
-
Vercel Release Bot authored
-
Niklas Mischkulnig authored
-
Benjamin Woodruff authored
chore(turbo-tasks): Remove argument resolution from backends, only allow local-task-based argument resolution (#75214) Removes the (now default) feature flag for local task argument resolution. With local-task based argument resolution in place, we don't need to support argument resolution in the backend, which simplifies some code.
-
Randolf J authored
### What? NextJS should not throw `The "use server" file can only export async functions.` for type exports. ### Why? Because type exports are just a compile time abstraction.
-
Tobias Koppers authored
### What? * change turbopack-cli to new backend * avoid transient tasks in turbopack-cli
-
Hendrik Liebau authored
The `"use cache"` directive is not compatible with the Edge runtime. When the `experimental.useCache` flag is enabled, we now emit a build error for pages with `export const runtime = 'edge'`. This is analogous to using the `experimental.dynamicIO` flag. The other route segment configs that are forbidden when `dynamicIO` is enabled, are currently still allowed for `useCache`: - `dynamicParams` - `dynamic` - `fetchCache` - `revalidate`
-
Niklas Mischkulnig authored
-
Hendrik Liebau authored
Fixes all issues related to the current export statement validation for app router pages, as highlighted in #75277.
-
Jiachi Liu authored
-
Hendrik Liebau authored
The issues with the current export statement validation for app router pages are documented in the added fixtures, see also the inline comments in the diff.
-
Lee Robinson authored
Co-authored-by:Jiachi Liu <inbox@huozhi.im>
-
Donny/강동윤 authored
### What? Reduce allocations for SWC plugins, by applying https://github.com/swc-project/plugins/pull/393 ### Why? Those are actually noop so they should not allocate too much.
-
Donny/강동윤 authored
### What? It **removes** needless allocations. ### Why? To avoid allocating memory needlessly.
-
Tobias Koppers authored
### What? Chunk async loaders together with normal chunk items to avoid multiple chunkings and small chunks
-
Sebastian "Sebbie" Silbermann authored
Allows us to remove some `any` casts and `@ts-expect-error`.
-
Jiwon Choi authored
-
Vercel Release Bot authored
-
Donny/강동윤 authored
### What? Apply https://github.com/swc-project/plugins/pull/392 ### Why? We should not create enormous amount of memory operations needlessly.
-
Niklas Mischkulnig authored
**View diff without whitespace changes!** This brings the number of `AssetIdent::path` calls down from 5M to 100k ``` ### canary (de984c2695b): ~16,60GB 582.76s user 109.72s system 815% cpu 1:24.95 total 583.30s user 108.45s system 805% cpu 1:25.86 total ### plus mischnic/client-manifest-optimization: still ~16,60GB 547.70s user 105.24s system 811% cpu 1:20.48 total 550.16s user 108.14s system 782% cpu 1:24.10 total ```
-
Tobias Koppers authored
### What? * use raw_read_dir to avoid creating many FileSystemPaths * use raw_read_dir in resolving
-
Tobias Koppers authored
That only causes incomplete spans in the tracing when the Span is cached.
-
Tobias Koppers authored
<!-- Thanks for opening a PR! Your contribution is much appreciated. To make sure your PR is handled as smoothly as possible we request that you follow the checklist sections below. Choose the right checklist for the change(s) that you're making: ## For Contributors ### Improving Documentation - Run `pnpm prettier-fix` to fix formatting issues before opening the PR. - Read the Docs Contribution Guide to ensure your contribution follows the docs guidelines: https://nextjs.org/docs/community/contribution-guide ### Adding or Updating Examples - The "examples guidelines" are followed from our contributing doc https://github.com/vercel/next.js/blob/canary/contributing/examples/adding-examples.md - Make sure the linting passes by running `pnpm build && pnpm lint`. See https://github.com/vercel/next.js/blob/canary/contributing/repository/linting.md ### Fixing a bug - Related issues linked using `fixes #number` - Tests added. See: https://github.com/vercel/next.js/blob/canary/contributing/core/testing.md#writing-tests-for-nextjs - Errors have a helpful link attached, see https://github.com/vercel/next.js/blob/canary/contributing.md ### Adding a feature - Implements an existing feature request or RFC. Make sure the feature request has been accepted for implementation before opening a PR. (A discussion must be opened, see https://github.com/vercel/next.js/discussions/new?category=ideas) - Related issues/discussions are linked using `fixes #number` - e2e tests added (https://github.com/vercel/next.js/blob/canary/contributing/core/testing.md#writing-tests-for-nextjs) - Documentation added - Telemetry added. In case of a feature if it's used or not. - Errors have a helpful link attached, see https://github.com/vercel/next.js/blob/canary/contributing.md ## For Maintainers - Minimal description (aim for explaining to someone not on the team to understand the PR) - When linking to a Slack thread, you might want to share details of the conclusion - Link both the Linear (Fixes NEXT-xxx) and the GitHub issues - Add review comments if necessary to explain to the reviewer the logic behind a change ### What? ### Why? ### How? Closes NEXT- Fixes # -->
-
vercel-release-bot authored
-
Jiachi Liu authored
-
- 23 Jan, 2025 7 commits
-
-
Hendrik Liebau authored
-
vercel-release-bot authored
-
Jiwon Choi authored
test: skip dev/deploy testing for `test/e2e/app-dir/segment-cache/revalidation/segment-cache-revalidation.test.ts` (#75249) ### Why? Since this test creates a mock server to fetch before creating a Next.js instance, it will fail on the deployment test due to a mismatch in the location of the server and the instance. The test did try to skip when deployed but it was after the `createNext` logic, which triggered the deployment before skipping. ### How? Move the skip before any `beforeAll` logic like creating an instance. --------- Co-authored-by:Zack Tanner <1939140+ztanner@users.noreply.github.com>
-
Jiwon Choi authored
Enable the new UI for the CI testings of existing redbox tests. There are several changes made to let the test pass, including backporting changes to the old UI or removing one from the new, and are as follows: - Added back `|` after line number in code frame ([link](https://github.com/vercel/next.js/pull/74935#discussion_r1925725941)) - Removed unnecessary `@` in Terminal component ([link](https://github.com/vercel/next.js/pull/74935#discussion_r1925727156)) - Set open overlay default value to `true` in Pages Router ([link](https://github.com/vercel/next.js/pull/74935#discussion_r1925728104)) - Backport displaying the first first-party call stack frame to the CallStack component ([link](https://github.com/vercel/next.js/pull/74935#discussion_r1925731313)) - Move the devTools component back to the error boundary ([link](https://github.com/vercel/next.js/pull/74935#discussion_r1925732706)) - Was moved out at https://github.com/vercel/next.js/pull/74999#discussion_r1922998021 Closes NDX-674 Closes NDX-687
-
Jiwon Choi authored
Moved `should not cause error when removing loading.js` test to flaky manifest. x-ref: [Flakiness Metrics](https://app.datadoghq.com/ci/test-runs?query=test_level%3Atest%20env%3Aci%20%40git.repository.id%3Agithub.com%2Fvercel%2Fnext.js%20%40test.service%3Anextjs%20%40test.status%3Afail%20%40test.name%3A%22app%20dir%20HMR%20should%20not%20cause%20error%20when%20removing%20loading.js%22&agg_m=count&agg_m_source=base&agg_t=count¤tTab=overview&eventStack=&fromUser=false&index=citest&start=1720817553405&end=1721422353405&paused=false)
-
HQidea authored
### Fixing a bug The code is located at https://github.com/vercel/next.js/blob/c33eb2b6505f6717a1a1eac10e193da9d9fa1ece/packages/next/src/server/lib/start-server.ts#L429 `typeof msg` should be compared to `'object'`. Otherwise, it makes no sense.
-
Donny/강동윤 authored
### Why? `core::ptr::drop_in_place` of `ReadRawVcFuture.turbo_task` is consuming too much time.
-