Parallelise and merge ci jobs, make sure rust dependencies are shared everywhere - #9752
Parallelise and merge ci jobs, make sure rust dependencies are shared everywhere#9752robert3005 wants to merge 2 commits into
Conversation
… everywhere Signed-off-by: Robert Kruszewski <github@robertk.io>
Merging this PR will regress 2 benchmarks
|
| Mode | Benchmark | BASE |
HEAD |
Efficiency | |
|---|---|---|---|---|---|
| ❌ | Simulation | random_i8[0.5] |
66.7 µs | 90.7 µs | -26.44% |
| ❌ | Simulation | allocate_drop_bytes[0] |
466 ns | 520.2 ns | -10.41% |
| ⚡ | Simulation | random_i8[0.8] |
97.9 µs | 69.2 µs | +41.44% |
| ⚡ | WallTime | mul_u64_nonnull_neon |
21.1 µs | 15.4 µs | +37.01% |
| ⚡ | WallTime | arrow_checked_add_u32_avx512[16384] |
21.4 µs | 17.6 µs | +21.61% |
| ⚡ | WallTime | arrow_checked_add_u32_avx2[16384] |
21.3 µs | 17.7 µs | +20.5% |
| ⚡ | WallTime | multiply_shapes_neon[(16384, PerRowPerRow)] |
20.2 µs | 17.3 µs | +16.32% |
| ⚡ | WallTime | mul_i64_nonnull_neon |
20.2 µs | 17.3 µs | +16.32% |
| 🆕 | WallTime | filtered_owned_i64_neon[AllValid] |
N/A | 4.5 µs | N/A |
| 🆕 | WallTime | filtered_owned_i64_neon[NineNullsInTen] |
N/A | 25.7 µs | N/A |
| 🆕 | WallTime | filtered_owned_i64_neon[OneNullInEight] |
N/A | 47.7 µs | N/A |
| 🆕 | WallTime | filtered_sink_i64_neon[AllValid] |
N/A | 4.8 µs | N/A |
| 🆕 | WallTime | filtered_sink_i64_neon[NineNullsInTen] |
N/A | 25.8 µs | N/A |
| 🆕 | WallTime | filtered_sink_i64_neon[OneNullInEight] |
N/A | 47.9 µs | N/A |
| 🆕 | WallTime | filtered_owned_i64_avx512[AllValid] |
N/A | 5.3 µs | N/A |
| 🆕 | WallTime | filtered_owned_i64_avx512[NineNullsInTen] |
N/A | 37.3 µs | N/A |
| 🆕 | WallTime | filtered_owned_i64_avx512[OneNullInEight] |
N/A | 58.7 µs | N/A |
| 🆕 | WallTime | filtered_sink_i64_avx512[AllValid] |
N/A | 5.3 µs | N/A |
| 🆕 | WallTime | filtered_sink_i64_avx512[NineNullsInTen] |
N/A | 37.3 µs | N/A |
| 🆕 | WallTime | filtered_sink_i64_avx512[OneNullInEight] |
N/A | 58 µs | N/A |
| ... | ... | ... | ... | ... | ... |
ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.
Tip
Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.
Comparing rk/parallelci (a53eae5) with develop (141b4bf)2
Footnotes
-
206 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. ↩
-
No successful run was found on
develop(78f4d38) during the generation of this report, so 141b4bf was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩
Signed-off-by: Robert Kruszewski <github@robertk.io>
1 participant
Leverage github's parallel steps to speed up our CI, merge steps where majority
of time was compilation