[robert3005]

Replace constant-set membership comparison/OR chains with prepared, single-pass probes

[codspeed]

Merging this PR will regress 12 benchmarks

⚠️ 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.

⚠️ 2 benchmarks measured no execution time

Nothing ran under measurement, usually because the compiler removed the code under test. These results are not comparable, so they count as unchanged.

Preventing compiler optimizations

• bench_compare_sliced_dict_primitive[(3333, 10000)]
• density_sweep_dense_runs[0.9]

⚠️ 8 benchmarks spent significant time in system calls

System calls cannot be consistently instrumented, so they are not included in the measure, which understates the real cost. Please switch to the Walltime instrument to accurately measure system calls.

Measurement and system calls

• random_i128[0.01]
• decompress[u16, (4000, 256)]
• take_varbin
• multipolygon_mixed_validity
• ilike_contains
• filter_run_end[len=4096_run=16_density=0.1]
• bench_compare_sliced_dict_primitive[(3333, 10000)]
• density_sweep_dense_runs[0.9]

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

⚡ 7 improved benchmarks
❌ 12 regressed benchmarks
✅ 2200 untouched benchmarks

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	optimize_lookup_predicate[ids=192, shape=in_list]	51 µs	76.7 µs	-33.44%
❌	Simulation	optimize_lookup_predicate[ids=128, shape=in_list]	51.6 µs	74 µs	-30.21%
❌	Simulation	optimize_lookup_predicate[ids=64, shape=in_list]	51.5 µs	71.4 µs	-27.83%
❌	Simulation	optimize_lookup_predicate[ids=16, shape=in_list]	52.4 µs	70.2 µs	-25.41%
❌	Simulation	optimize_lookup_predicate[ids=256, shape=in_list]	79.6 µs	106.1 µs	-24.97%
❌	Simulation	optimize_lookup_predicate[ids=1, shape=in_list]	75.3 µs	90.7 µs	-17.07%
❌	Simulation	sum_v2_i64	194.2 µs	224.9 µs	-13.63%
❌	Simulation	sum_i64	194.4 µs	224.7 µs	-13.48%
❌	Simulation	random_i8[0.01]	108.4 µs	123.9 µs	-12.5%
❌	WallTime	filter_avx2[64]	2.1 µs	2.4 µs	-12.23%
❌	WallTime	filter_avx2[80]	2.2 µs	2.5 µs	-10.93%
❌ ⚠️	Simulation	random_i128[0.01]	7.2 µs	8 µs	-10.53%
⚡	Simulation	i64_random_chunked[256]	13,907.4 µs	226.4 µs	×61
⚡	Simulation	utf8_random[256]	12,973.6 µs	416.1 µs	×31
⚡	Simulation	i64_random[256]	3,928.8 µs	170.1 µs	×23
⚡	Simulation	nested_list_random[32]	6.7 ms	1 ms	×6.4
⚡	WallTime	filter_avx512[80]	3.2 µs	2.3 µs	+38.85%
⚡	WallTime	filter_avx512[64]	2.9 µs	2.2 µs	+33.18%
⚡	WallTime	words_gather_dispatch_neon[65536]	2.3 µs	2 µs	+10.42%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing claude/friendly-knuth-dj6ed1-list-contains (4bb2b50) with develop (751e0af)}

[joseph-isaacs]

How do i extend this to work for new custom kernels?

I really think we want to make probe an array to allow reduce/execute rules?

[joseph-isaacs]

Why is this defined here and in scalar_fn?

[joseph-isaacs]

why is this used in two places

seems like both ConstantArray list_contains and list_contains scalar fn vtable share the constant heystack impl?

[joseph-isaacs]

shared with vortex-compute?

[joseph-isaacs]

not this

[joseph-isaacs]

Comments worth fixing, but otherall good

[robert3005]

I will have a follow up to convert the existing trait into an Array