Replace MaybeUninit transmutes with write_copy_of_slice - #9814
Replace MaybeUninit transmutes with write_copy_of_slice#9814robert3005 wants to merge 3 commits into
Performance Regression: -6.36%
⚠️ 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.
⚡ 4 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 | |
|---|---|---|---|---|---|
| ❌ | WallTime | arrow_checked_add_u32_neon[16384] |
12.3 µs | 20.3 µs | -39.57% |
| ❌ | Simulation | random_i8[0.5] |
67.8 µs | 89.5 µs | -24.24% |
| ❌ | WallTime | filtered_owned_i64_avx512[OneNullInEight] |
22.5 µs | 26.1 µs | -14.08% |
| ❌ | WallTime | filtered_sink_i64_neon[NineNullsInTen] |
14.7 µs | 16.8 µs | -12.56% |
| ❌ | WallTime | words_gather_scalar_avx2[65536] |
8.2 µs | 9.4 µs | -12.15% |
| ❌ | WallTime | filtered_owned_i64_avx512[NineNullsInTen] |
13.5 µs | 15.2 µs | -11.24% |
| ❌ | Simulation | random_i128[0.8] |
170.9 µs | 191.4 µs | -10.71% |
| ❌ | Simulation | patterns_i128[Runs] |
106.8 µs | 119.6 µs | -10.67% |
| ❌ | WallTime | dict_canonicalize_gt_u8_neon[1000000] |
486.9 µs | 542.4 µs | -10.24% |
| ⚡ | WallTime | dict_canonicalize_gt_u8_avx2[16000000] |
8.9 ms | 6.8 ms | +31.41% |
| ⚡ | Simulation | random_i16[0.95] |
95.6 µs | 75.6 µs | +26.55% |
| ⚡ | WallTime | filtered_owned_i64_avx2[OneNullInEight] |
25.8 µs | 22 µs | +17.44% |
| ⚡ | Simulation | allocate_drop_arrow[0] |
456.9 ns | 402.7 ns | +13.45% |
Tip
Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.
Comparing rk/uninitwriteslice (2720cef) with develop (8e2aa05)
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. ↩