perf: optimize create chunk assets

[LingyuCoder]

Summary

Round 6 for rust@create_chunk_assets reached the requested +5% target.

• skip per-asset chunk_asset hook dispatch when the hook has no actual taps, including JS interceptor registrations that resolve to no taps
• keep full-build render results as a linear Vec and avoid building ChunkRenderArtifact/FxHashMap unless incremental chunk-asset caching is enabled
• collect full-build chunks into a Vec instead of hashing them into a set before render-manifest dispatch; incremental updates still dedupe affected chunks
• split the no-chunk_asset fast path from the hooked path so the common path avoids a per-chunk hook-state branch
• preserve the previous non-fatal behavior for interceptor registration errors by falling back to the hooked path instead of propagating probe errors
• skip empty diagnostics extension while processing chunk render results
• preallocate chunk render result storage before merging parallel render manifest output
• preallocate each chunk render manifest vector for the common small-manifest case
• keep the existing per-asset path when chunk_asset is observed by Rust or JS plugins

Expected performance mechanism

The create chunk assets stage can emit many chunk assets. In the common full-build no-listener path, this avoids hash-set construction for the chunk list, resolves chunk asset taps once for the pass, avoids an async hook call per emitted file, avoids full-build hash-map/artifact construction, reduces small Vec growth while collecting render manifests, skips empty diagnostics merging, batches chunk rendered/file bookkeeping per chunk, and keeps the hot no-hook loop separate from the compatibility hook loop.

Validation

CI-only workflow: local build, test, lint, and benchmark commands were not run. cargo fmt --all was run only to apply formatting after CI reported a rustfmt diff in an earlier round.

Round 6 head: 96029d1fd1c71b1361f3a5b80db73e5febb80df9.

GitHub Actions for this head passed:

• CI Diff: success
• CI-Dylint: success
• CI-Rust: success
• CI: success
• CI-Lint: success

CodSpeed for this head reported:

• rust@create_chunk_assets: 3.2 ms -> 3 ms, +5.14%
• js@collect imported identifiers: 222.4 us -> 228.6 us, -2.71%

CodSpeed also reported Walltime/runtime-environment warnings, so the exact percentages should be read with that context. Binary size-limit reported +12.00KB / +0.02% for this head.

[Copilot]

Pull request overview

This PR targets the create_chunk_assets compilation pass to reduce overhead when many assets are emitted, primarily by avoiding per-asset chunk_asset hook dispatch in the no-listener case and by reducing allocation churn when collecting parallel render results.

Changes:

• Preallocates storage for chunk render results before merging parallel render_manifest outputs.
• Adds a fast path that batches per-chunk “rendered/files bookkeeping” and skips per-asset chunk_asset hook calls when the hook is empty.
• Preserves the existing per-asset behavior when chunk_asset is observed (taps/interceptors present).

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[codspeed-hq]

Merging this PR will not alter performance

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

⚡ 1 improved benchmark
❌ 1 regressed benchmark
✅ 50 untouched benchmarks
⏩ 40 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	js@collect imported identifiers	222.4 µs	228.6 µs	-2.71%
⚡	Simulation	rust@create_chunk_assets	3.2 ms	3 ms	+5.14%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing codex/perf-create-chunk-assets (96029d1) with main (b04d9f4)}

Footnotes

1. 40 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

[github-actions]

📦 Binary Size-limit

Comparing 96029d1 to feat: support rspack magic comment prefix (#14323) by AsyncIter

❌ Size increased by 12.00KB from 62.60MB to 62.61MB (⬆️0.02%)

[github-actions]

Rsdoctor Bundle Diff Analysis

Found 5 projects in monorepo, 0 projects with changes.

📊 Quick Summary

Project	Total Size	Gzip Size	Change	Gzip Change
popular-libs	1.7 MB	551.3 KB	0	0
react-10k	5.6 MB	1.3 MB	0	0
react-1k	822.8 KB	218.3 KB	0	0
react-5k	2.7 MB	669.1 KB	0	0
ui-components	4.8 MB	1.4 MB	0	0

Generated by Rsdoctor GitHub Action

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 5de5eb2107

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".