Add associated outputs to RowFn kernels - #9548
2 benchmarks regressed
⚠️ Unknown Walltime execution environment detected
Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.
For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.
⚠️ Different runtime environments detected
Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.
⚡ 9 improved benchmarks
❌ 2 regressed benchmarks
✅ 1970 untouched benchmarks
⏩ 54 skipped benchmarks1
Warning
Please fix the performance issues or acknowledge them on CodSpeed.
Performance Changes
| Mode | Benchmark | BASE |
HEAD |
Efficiency | |
|---|---|---|---|---|---|
| ❌ | Simulation | chunked_varbin_opt_to_varbinview_builder[(10, 100)] |
558.2 µs | 666.6 µs | -16.26% |
| ❌ | Simulation | cold_misaligned[(16, 64)] |
378.6 µs | 429.6 µs | -11.88% |
| ⚡ | Simulation | subtract_shapes[(128, PerRowConstant)] |
90.5 µs | 75.2 µs | +20.38% |
| ⚡ | Simulation | multiply_shapes[(128, PerRowConstant)] |
91.7 µs | 76.6 µs | +19.82% |
| ⚡ | Simulation | add_shapes[(128, PerRowNullableConstant)] |
103.9 µs | 87 µs | +19.44% |
| ⚡ | Simulation | add_shapes[(128, PerRowConstant)] |
90.7 µs | 76 µs | +19.22% |
| ⚡ | Simulation | multiply_shapes[(128, ConstantPerRow)] |
91.9 µs | 77.5 µs | +18.6% |
| ⚡ | Simulation | multiply_shapes[(128, PerRowNullableConstant)] |
105.2 µs | 89.1 µs | +18.17% |
| ⚡ | Simulation | subtract_shapes[(128, ConstantPerRow)] |
90.1 µs | 76.4 µs | +18.04% |
| ⚡ | Simulation | add_shapes[(128, ConstantPerRow)] |
89.8 µs | 76.1 µs | +18.02% |
| ⚡ | Simulation | subtract_shapes[(128, PerRowNullableConstant)] |
102.9 µs | 88.2 µs | +16.74% |
Tip
Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.
Comparing ct/row-fn-associated-output (4aa169c) with develop (332ae0b)
Footnotes
-
54 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. ↩