Execute tensor L2 norm with RowFn - #9768
Performance Regression: -5.14%
⚠️ 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.
⚡ 5 improved benchmarks
❌ 9 regressed benchmarks
✅ 2184 untouched benchmarks
⏩ 218 skipped benchmarks1
Warning
Please fix the performance issues or acknowledge them on CodSpeed.
Performance Changes
| Mode | Benchmark | BASE |
HEAD |
Efficiency | |
|---|---|---|---|---|---|
| ❌ | Simulation | column_x_extension_constant[32] |
170.5 µs | 234.5 µs | -27.3% |
| ❌ | Simulation | column_x_extension_constant[256] |
414.6 µs | 563.8 µs | -26.46% |
| ❌ | Simulation | random_i8[0.5] |
71.5 µs | 95.4 µs | -24.98% |
| ❌ | Simulation | column_x_column[256] |
88.9 µs | 112.5 µs | -20.97% |
| ❌ | Simulation | column_x_column[32] |
94.4 µs | 117.7 µs | -19.78% |
| ❌ | Simulation | decompress[u64, (4000, 1024)] |
71.9 µs | 87.3 µs | -17.58% |
| ❌ | Simulation | column_x_constant[256] |
591.1 µs | 711.5 µs | -16.92% |
| ❌ | Simulation | column_x_constant[32] |
392.9 µs | 448.2 µs | -12.34% |
| ❌ | Simulation | column_x_extension_constant[2] |
247.8 µs | 276.8 µs | -10.48% |
| ⚡ | Simulation | nullable[2] |
632.3 µs | 416.6 µs | +51.8% |
| ⚡ | Simulation | non_nullable[2] |
631.2 µs | 419.1 µs | +50.59% |
| ⚡ | Simulation | random_i16[0.95] |
99 µs | 81 µs | +22.2% |
| ⚡ | Simulation | allocate_drop_arrow[0] |
456.9 ns | 402.7 ns | +13.45% |
| ⚡ | Simulation | allocate_drop_bytes[0] |
575.7 ns | 521.6 ns | +10.39% |
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-tensor-l2-v2 (b84f3f7) with develop (7286721)
Footnotes
-
218 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. ↩