Skip to content

Latest commit

 

History

History
19 lines (16 loc) · 1.05 KB

evaluation.md

File metadata and controls

19 lines (16 loc) · 1.05 KB

Shared benchmark results

Raw corpus runs are under corpus/results/ and raw micro runs are under micro/results/. Test artifacts are under tests/results/. These directories retain the original result.json records; the reorganization does not alter them.

Question Current interpretation
kprog native execution kprog evaluation
kinsn inline kfuncs kinsn evaluation
Speculative ReJIT ReJIT evaluation
Cross-runtime micro characterization micro benchmark status
Corpus workload selection workload tuning log

Run make micro or make corpus from the repository root to create new results. For exact datasets, sample counts, and commands used in a claim, follow the corresponding project evaluation document. Dated investigations are retained in the shared archive.