Conversation
Merging this PR will regress 1 benchmark
|
| Mode | Benchmark | BASE |
HEAD |
Efficiency | |
|---|---|---|---|---|---|
| ❌ | WallTime | mul_i32_nonnull_avx512 |
7.1 µs | 7.9 µs | -10.38% |
| ⚡ | WallTime | filtered_sink_i64_avx2[OneNullInEight] |
25.8 µs | 23.1 µs | +11.35% |
| ⚡ | WallTime | mul_u32_nonnull_avx512 |
6.3 µs | 5.6 µs | +11.25% |
Tip
Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.
Comparing mk/bitpacked-stack-01-wire-boundary (9681199) with develop (c6678f3)
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. ↩
Signed-off-by: "Matt Katz" <mhkatz97@gmail.com>
950de57 to
9681199
Compare
Move serialization and deserialization of
fastlanes.bitpackedintoBitPackedPlugin, shared with the external-patches adapter. The plugin owns the protobuf metadata and wire validation.The VTable retains only the required error-returning serde methods and in-memory array validation.