perf-cache-cur-kind - Branch - oxc-project/oxc

perf(parser): cache cur_kind() to eliminate redundant calls

#14411Merged

Comparing

perf-cache-cur-kind

(

8f056ad

) with

main

(

2b6a3b4

)

Untouched: 37

Benchmarks

Passed

parser[RadixUIAdoptionSection.jsx]

tasks/benchmark/benches/parser.rs::parser::bench_parser

+1%

87.7 µs87.2 µs

parser[react.development.js]

tasks/benchmark/benches/parser.rs::parser::bench_parser

1.3 ms1.3 ms

parser[binder.ts]

tasks/benchmark/benches/parser.rs::parser::bench_parser

3.3 ms3.3 ms

semantic[react.development.js]

tasks/benchmark/benches/semantic.rs::semantic::bench_semantic

1.6 ms1.6 ms

formatter[react.development.js]

tasks/benchmark/benches/formatter.rs::formatter::bench_formatter

12.1 ms12.1 ms

formatter[RadixUIAdoptionSection.jsx]

tasks/benchmark/benches/formatter.rs::formatter::bench_formatter

583.2 µs582.7 µs

codegen[RadixUIAdoptionSection.jsx]

tasks/benchmark/benches/codegen.rs::codegen::bench_codegen

119.1 µs119 µs

transformer[react.development.js]

tasks/benchmark/benches/transformer.rs::transformer::bench_transformer

714.8 µs714.3 µs

semantic[RadixUIAdoptionSection.jsx]

tasks/benchmark/benches/semantic.rs::semantic::bench_semantic

71.3 µs71.3 µs

codegen[react.development.js]

tasks/benchmark/benches/codegen.rs::codegen::bench_codegen

1.8 ms1.8 ms

transformer[binder.ts]

tasks/benchmark/benches/transformer.rs::transformer::bench_transformer

1.7 ms1.7 ms

codegen[binder.ts]

tasks/benchmark/benches/codegen.rs::codegen::bench_codegen

4 ms4 ms

linter[react.development.js]

tasks/benchmark/benches/linter.rs::linter::bench_linter

33.1 ms33.1 ms

isolated-declarations[vue-id.ts]

tasks/benchmark/benches/transformer.rs::transformer::bench_isolated_declarations

56.4 ms56.3 ms

linter[cal.com.tsx]

tasks/benchmark/benches/linter.rs::linter::bench_linter

731.5 ms731.3 ms

formatter[cal.com.tsx]

tasks/benchmark/benches/formatter.rs::formatter::bench_formatter

211.9 ms211.8 ms

formatter[binder.ts]

tasks/benchmark/benches/formatter.rs::formatter::bench_formatter

23.8 ms23.8 ms

parser[cal.com.tsx]

tasks/benchmark/benches/parser.rs::parser::bench_parser

26.6 ms26.6 ms

linter[binder.ts]

tasks/benchmark/benches/linter.rs::linter::bench_linter

89 ms89 ms

mangler[cal.com.tsx]

tasks/benchmark/benches/minifier.rs::minifier::bench_mangler

2.8 ms2.8 ms

lexer[cal.com.tsx]

tasks/benchmark/benches/lexer.rs::lexer::bench_lexer

5.3 ms5.3 ms

estree[checker.ts]

tasks/benchmark/benches/parser.rs::parser::bench_estree

104.5 ms104.5 ms

lexer[RadixUIAdoptionSection.jsx]

tasks/benchmark/benches/lexer.rs::lexer::bench_lexer

19.8 µs19.8 µs

semantic[cal.com.tsx]

tasks/benchmark/benches/semantic.rs::semantic::bench_semantic

26.4 ms26.4 ms

lexer[binder.ts]

tasks/benchmark/benches/lexer.rs::lexer::bench_lexer

863.2 µs863.3 µs

lexer[react.development.js]

tasks/benchmark/benches/lexer.rs::lexer::bench_lexer

354.2 µs354.3 µs

mangler[react.development.js]

tasks/benchmark/benches/minifier.rs::minifier::bench_mangler

262.9 µs262.9 µs

linter[RadixUIAdoptionSection.jsx]

tasks/benchmark/benches/linter.rs::linter::bench_linter

1.6 ms1.6 ms

transformer[RadixUIAdoptionSection.jsx]

tasks/benchmark/benches/transformer.rs::transformer::bench_transformer

134.4 µs134.4 µs

codegen[cal.com.tsx]

tasks/benchmark/benches/codegen.rs::codegen::bench_codegen

34.1 ms34.2 ms

semantic[binder.ts]

tasks/benchmark/benches/semantic.rs::semantic::bench_semantic

4.1 ms4.1 ms

minifier[binder.ts]

tasks/benchmark/benches/minifier.rs::minifier::bench_minifier

4 ms4 ms

minifier[react.development.js]

tasks/benchmark/benches/minifier.rs::minifier::bench_minifier

2.5 ms2.5 ms

minifier[cal.com.tsx]

tasks/benchmark/benches/minifier.rs::minifier::bench_minifier

35.8 ms35.8 ms

mangler[binder.ts]

tasks/benchmark/benches/minifier.rs::minifier::bench_mangler

745.3 µs745.9 µs

transformer[cal.com.tsx]

tasks/benchmark/benches/transformer.rs::transformer::bench_transformer

26.4 ms26.4 ms

mangler[RadixUIAdoptionSection.jsx]

tasks/benchmark/benches/minifier.rs::minifier::bench_mangler

13.1 µs13.1 µs

Commits

Click on a commit to change the comparison range

Base

main

2b6a3b4

+0.04%

perf(parser): cache cur_kind() to eliminate redundant calls (#14411) ## Summary Optimization to reduce redundant token kind extraction in hot paths by caching `cur_kind()` results before compound checks. ## Problem - Compound checks like `self.at(X) || self.at(Y)` call `cur_kind()` twice - Pattern `self.at(close) || self.has_fatal_error()` calls `cur_kind()` twice since `has_fatal_error()` internally calls it - Token kind extraction involves bit-packing operations: `((self.0 >> 64) & 0xFF) as u8` ## Solution Cache `cur_kind()` in a local variable before compound checks to eliminate redundant calls. ## Changes - **`cursor.rs:parse_delimited_list()`**: Hot loop parsing arrays, objects, parameters - reduced from 2-3 calls per iteration to 1 - **`arrow.rs:68`**: Cached kind for `LParen`/`LAngle` check - **`module.rs:101`**: Cached kind for `LCurly`/`Star` check - **`types.rs`**: Fixed 3 locations (lines 91, 558, 1188) with double `at()` calls ## Impact Every JavaScript array, object literal, function parameter list, and TypeScript type now extracts the token kind fewer times during parsing. 🤖 Generated with [Claude Code](https://claude.com/claude-code)

8f056ad

1 month ago

by Boshen

Home Terms Privacy Docs