Skip to content

1.1 release#20

Merged
Compaile merged 21 commits intomainfrom
v103_performance
Sep 29, 2025
Merged

1.1 release#20
Compaile merged 21 commits intomainfrom
v103_performance

Conversation

@Compaile
Copy link
Copy Markdown
Owner

No description provided.

root and others added 21 commits June 8, 2025 19:51
- Add CMake option BUILD_BENCHMARK (default OFF) to build benchmark
- Implement ctrack_benchmark with four measurement categories:
  1. Accuracy: Measures how accurately CTRACK records function timings
  2. Overhead: Compares performance with/without CTRACK enabled
  3. Memory: Tracks memory usage per recorded event
  4. Calculation time: Measures time to process results
- Support configurable event count (default 50M) and thread count
- Add baseline recording (--record-baseline) and comparison (--compare-baseline)
- Benchmark uses nested functions with predictable timing patterns
- Includes multi-threaded stress testing capabilities

Usage: ctrack_benchmark [--events N] [--threads N] [--record-baseline] [--compare-baseline]

This benchmark will help track performance regressions and improvements
across different platforms and CTRACK versions.
- Add comprehensive Developer Guide section to README.md
- Document all benchmark features including baseline comparison
- Provide best practices for running and interpreting benchmarks
- Add CLAUDE.local.md to .gitignore for local documentation
… add accuracy error measurement and enhance baseline data structure
…oved performance and adjust event capacity reservation strategy to optimize memory usage.
- Implemented tests for summary row validation, including basic and multiple function scenarios.
- Added detailed statistics validation tests to ensure accurate tracking of function performance.
- Created tests for result meta information and settings filtering to verify correct behavior under various configurations.
- Included tests for statistical calculations, including variance, percentiles, and outlier handling.
- Ensured consistency across multiple calls and edge cases, such as single executions and extreme values.
- Validated statistical stability with large sample sizes and progressive variance analysis.
@Compaile Compaile changed the title V103 performance 1.1 release Sep 29, 2025
@Compaile Compaile marked this pull request as ready for review September 29, 2025 10:22
@Compaile Compaile merged commit 6dfa9b0 into main Sep 29, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant