[connortsui20]

Summary

• Tracking Issue: Tracking Issue: Define the RowFn API #9129

Separates the borrowed-decoding extension point from the internal execution changes in #9979. Native primitive inputs can share their canonical buffers without cloning the array handle.

Targets develop and can merge independently.

Changes

Defines equivalent values, row counts, retained storage, and decoder guarantees for owned and borrowed decoding. New cases cover retained storage for the default decoder, canonical primitives, and encoded primitives.

API Changes

Adds InputElement::decode_ref(&ArrayRef, &mut ExecutionCtx) with a default that clones and delegates to decode. Existing element implementations need no changes. RowFn authors keep the existing visitor API.

The Vortex test suite was not run locally. Focused native measurements of the combined migration were run separately during the compiler investigation.

[codspeed]

Merging this PR will regress 4 benchmarks

3 warnings, 1 note

⚠️ Walltime results may be unreliable

Some benchmarks used the Walltime instrument outside CodSpeed Macro Runners, which can make the results inconsistent.

Learn more about Macro Runners

See affected benchmarks

⚠️ 17 benchmarks are dominated by system calls

System calls cannot be simulated consistently and are left out of the measure. The reported value understates the real cost. Switch to the Walltime instrument to measure them.

Learn more about system calls

See affected benchmarks

⚠️ Base and head ran on different runner types

Some benchmarks with significant performance changes were compared across different runner types, so the changes might not come from the code. Run base and head on the same runner type, or use CodSpeed Macro Runners for more stable results.

Learn more about Macro Runners

Open the report in CodSpeed to investigate

ℹ️ 40 benchmarks have no flamegraph

The flamegraph wasn't generated: the symbols of the profiling trace could not be resolved, because no information about the language frames was collected.

See affected benchmarks

⚡ 8 improved benchmarks
❌ 4 regressed benchmarks
✅ 2396 untouched benchmarks
⏩ 12 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	CPU Simulation	bench_compare_sliced_dict_primitive[(2000, 10000)]	13 µs	67.3 µs	-80.69%
❌	CPU Simulation	density_sweep_dense_runs[0.001]	44.5 µs	61.3 µs	-27.39%
❌	Walltime	bitpack_blocked_compress_avx512	4.9 µs	5.6 µs	-13.61%
❌	CPU Simulation	random_i128[0.01]	5.3 µs	6.1 µs	-12.69%
⚡	Walltime	filter_neon[8]	1,748 ns	936 ns	+86.75%
⚡	Walltime	threshold_neon[8]	2.4 µs	1.4 µs	+76.52%
⚡	Walltime	threshold_neon[64]	4.3 µs	2.6 µs	+68.32%
⚡	Walltime	filter_neon[64]	2.7 µs	1.8 µs	+53.61%
⚡	Walltime	threshold_neon[80]	4.5 µs	3.4 µs	+32.19%
⚡	Walltime	filter_neon[80]	3.2 µs	2.6 µs	+23.05%
⚡	Walltime	bitpack_blocked_compress_avx2	7.6 µs	6.7 µs	+13.18%
⚡	CPU Simulation	test_scan_filter[int]	3.4 ms	3 ms	+12.6%

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-cleanup/borrowed-decode (aa5d0ec) with develop (dbaf5f8)}

Footnotes

1. 12 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. ↩