[DO NOT MERGE] Focused RowFn take_filter CodSpeed ablation

[connortsui20]

Runs only cargo codspeed run --bench take_filter in a pull-request context.

The framework-only revision 0a0ad0db1 was clean at 232.542 µs versus 233.737 µs for develop. The direct child 89fd28bc1, which moves primitive numeric execution to RowFn, reproduced the representative regression at 279.491 µs, or 16.37% slower.

This temporary PR has finished its ablation. Do not merge it.

[codspeed-hq]

Merging this PR will degrade performance by 13.71%

⚡ 2 improved benchmarks
❌ 25 regressed benchmarks
✅ 1917 untouched benchmarks
🆕 1 new benchmark
⏩ 85 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	compact_sliced[(16384, 10)]	1.4 ms	1.9 ms	-26.51%
❌	Simulation	compact_sliced[(4096, 10)]	385.6 µs	514.7 µs	-25.09%
❌	Simulation	multipolygon_random_nulls	942.4 µs	1,216.9 µs	-22.56%
❌	Simulation	multipolygon_mixed_validity	1.2 ms	1.5 ms	-18.73%
❌	Simulation	multipolygon_non_nullable	1.2 ms	1.5 ms	-18.42%
❌	Simulation	cold_misaligned[(64, 256)]	4.4 ms	5.2 ms	-16.24%
❌	Simulation	take_filter_list_small_uncached_random_mask_random_indices[256, 10]	233.7 µs	276.9 µs	-15.59%
❌	Simulation	take_filter_list_random_mask_sequential_indices[256, 50]	241.8 µs	283.9 µs	-14.84%
❌	Simulation	take_filter_list_slice_mask_random_indices[256, 50]	239.1 µs	280.3 µs	-14.71%
❌	Simulation	take_filter_list_small_random_mask_random_indices[256, 10]	232.8 µs	272.8 µs	-14.66%
❌	Simulation	take_filter_list_random_mask_random_indices[256, 50]	245.9 µs	287.9 µs	-14.58%
❌	Simulation	take_filter_list_small_random_mask_random_indices[768, 10]	242 µs	283.1 µs	-14.51%
❌	Simulation	take_filter_list_random_mask_sequential_indices[768, 50]	236.8 µs	276.8 µs	-14.45%
❌	Simulation	take_filter_list_slice_mask_random_indices[768, 50]	240.9 µs	281.5 µs	-14.41%
❌	Simulation	take_filter_list_random_mask_random_indices[768, 50]	256.1 µs	298.8 µs	-14.31%
❌	Simulation	take_filter_list_nullable_random_mask_random_indices[768, 50]	267.4 µs	310.7 µs	-13.93%
❌	Simulation	take_filter_list_slice_mask_sequential_indices[768, 50]	236.8 µs	274.6 µs	-13.77%
❌	Simulation	take_filter_list_small_uncached_random_mask_random_indices[768, 10]	242.2 µs	280.5 µs	-13.64%
❌	Simulation	take_filter_list_nullable_random_mask_random_indices[256, 50]	256.9 µs	297 µs	-13.52%
❌	Simulation	take_filter_list_slice_mask_sequential_indices[256, 50]	237.4 µs	274.5 µs	-13.52%
...	...	...	...	...	...

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 ct/row-fn-codspeed-take-filter (89fd28b) with develop (66d096b)}

Footnotes

1. 85 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. ↩