perf(ast): mark AstKind::span as #[inline]

[overlookmotel]

Now that AstKind's variants are all structs (no enums any more), and #20584 has ensured that Span is in a consistent position in all structs, AstKind::span should boil down to a single operation - a pointer read. Therefore, mark it #[inline].

Note: Span was already in a consistent position prior to #20584, but that was more by accident than anything else. Now it's guaranteed. But this change won't make a difference to perf, it just guards against a perf regression in future.

[overlookmotel]

• refactor(ast_tools): simplify struct field reordering #20587
• perf(ast): mark AstKind::address as #[inline] #20586
• perf(ast): mark AstKind::span as #[inline] #20585 👈 (View in Graphite)
• perf(ast): place NodeId field after Span in structs #20584
• refactor(ast): re-order layout assertions in memory order #20583
• refactor(ast_tools): convert layout calculation to a struct with methods #20582
• fix(ast_tools): remove NonMax32 primitive #20581
• fix(ast_tools): correctly parse define_index_type! macros #20580
• perf(ast_tools): reduce data passing between functions #20579
• perf(ast_tools): parse files in parallel #20578
• refactor(ast_tools): search all crates which depend on oxc_ast_macros #20577
• main

---

How to use the Graphite Merge Queue

Add either label to this PR to merge it via the merge queue:

• 0-merge - adds this PR to the back of the merge queue
• hotfix - for urgent changes, fast-track this PR to the front of the merge queue

You must have a Graphite account in order to use the merge queue. Sign up using this link.

An organization admin has enabled the Graphite Merge Queue in this repository.

Please do not merge from GitHub as this will restart CI on PRs being processed by the merge queue.

This stack of pull requests is managed by Graphite. Learn more about stacking.

[codspeed-hq]

Merging this PR will not alter performance

✅ 53 untouched benchmarks
⏩ 3 skipped benchmarks¹

---

_{Comparing om/02-15-perf_ast_mark_astkind_span_as_inline_ (c9edd39) with om/02-15-perf_ast_place_nodeid_field_after_span_in_structs (b258884)²}

Footnotes

1. 3 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. ↩

2. No successful run was found on om/02-15-perf_ast_place_nodeid_field_after_span_in_structs (2064ade) during the generation of this report, so 9b52103 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

[Copilot]

Pull request overview

This PR adds an #[inline] hint to AstKind’s GetSpan::span implementation to help prevent future performance regressions now that AST node span fields are guaranteed to be laid out consistently.

Changes:

• Add documentation explaining why AstKind::span should be cheap (consistent span field placement).
• Add #[inline] to the GetSpan for AstKind<'_> span() implementation in the generator and regenerated output.

Reviewed changes

Copilot reviewed 1 out of 2 changed files in this pull request and generated no comments.

File	Description
tasks/ast_tools/src/generators/ast_kind.rs	Updates the code generator to emit docs + #[inline] on AstKind’s GetSpan::span.
crates/oxc_ast/src/generated/ast_kind.rs	Regenerates AstKind to include the new docs + #[inline] on span().

[graphite-app]

Merge activity

• Mar 20, 10:15 PM UTC: overlookmotel added this pull request to the Graphite merge queue.
• Mar 20, 10:30 PM UTC: Merged by the Graphite merge queue.