Add GitHub Actions workflow for benchmarks#98
Conversation
Co-authored-by: dazinator <3176632+dazinator@users.noreply.github.com>
Co-authored-by: dazinator <3176632+dazinator@users.noreply.github.com>
|
@copilot any way we can get some visualisations of the benchmarks uploaded as artifacts like charts etc? Can benchmarkdotnet output that? |
Co-authored-by: dazinator <3176632+dazinator@users.noreply.github.com>
Co-authored-by: dazinator <3176632+dazinator@users.noreply.github.com>
Yes! I've added HTML and CSV exporters to the benchmark configuration. BenchmarkDotNet will now generate:
These will be included in the uploaded artifacts alongside the markdown reports. Commit: 70e0873 |
Co-authored-by: dazinator <3176632+dazinator@users.noreply.github.com>
Co-authored-by: dazinator <3176632+dazinator@users.noreply.github.com>
There was a problem hiding this comment.
Pull Request Overview
This PR adds a GitHub Actions workflow to automate benchmark execution for the DotNet.Glob project. The workflow enables running benchmarks on push/PR events or manually via workflow dispatch, addressing the need for automated performance testing.
Key Changes:
- Created a new GitHub Actions workflow that installs required .NET versions, builds and runs benchmarks with appropriate configurations
- Enhanced benchmark output by adding HTML and CSV exporters for better visualization and analysis of results
- Configured the workflow to handle project-specific requirements like disabling broken NuGet sources and assembly signing
Reviewed Changes
Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.
| File | Description |
|---|---|
| .github/workflows/benchmarks.yml | New workflow file that orchestrates benchmark execution with multi-version .NET setup, dependency management, and artifact uploads |
| src/DotNet.Glob.Benchmarks/Program.cs | Updated to configure benchmark exporters (HTML/CSV) and apply configuration to all benchmark runs |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
|
@copilot see pr bot feedback |
The bot's feedback about multiple frameworks doesn't apply here - the benchmarks project only targets
The hardcoded framework is correct and intentional since there's only one target framework in the project. |
Original prompt
💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.