[luketpeterson]

Adding sharing validator tool
Moving differential tool into validation dir where other validators can go

[github-actions]

Bench A/B vs base: done

job log · 2026-11-03 04:07:29 UTC

base b384d1b → head 534400c, 3 round(s), median of each run averaged; negative is faster

bench	cases	geomean	largest gain	largest loss	>5% faster	>5% slower
shakespeare	7	-1.9%	-6.4% shakespeare/shakespeare_sentences_val_count	+2.0% shakespeare/shakespeare_sentences_get	1	0
cities	5	-0.9%	-3.0% cities/cities_val_count_act	+0.1% cities/cities_get	0	0
sparse_keys	108	-0.4%	-6.0% sparse_iter/10	+6.5% sparse_meet/50	1	3
binary_keys	89	-0.8%	-3.6% binary_descend_until/500	+15.9% binary_insert/1600	0	1
superdense_keys	116	-0.4%	-12.4% superdense_zipper_step_iter/200	+14.3% superdense_drop_bench/200	4	1
act_paths	46	-0.8%	-7.1% shape_map_to_act_cata/random_long	+5.0% size_dense_paths_to_act/50000	2	0
zipper_head_owned	16	+0.9%	-2.6% zipper_head_owned/owned_head_read_creation	+12.4% zipper_head_owned/map_reader_val	0	2
product_zipper	4	-0.5%	-1.0% product_zipper/generic_pathmap_pathmap	+0.9% product_zipper/generic_act_act	0	0

15 case(s) moved more than 5%

bench	case	base	head	change
binary_keys	binary_insert/1600	155.43 µs	180.17 µs	+15.9%
superdense_keys	superdense_drop_bench/200	70 ns	80 ns	+14.3%
zipper_head_owned	zipper_head_owned/map_reader_val	2.75 µs	3.09 µs	+12.4%
superdense_keys	superdense_zipper_step_iter/200	1.69 µs	1.49 µs	-12.4%
act_paths	shape_map_to_act_cata/random_long	42.81 ms	39.79 ms	-7.1%
superdense_keys	superdense_drop_head/4000	28.93 µs	27.01 µs	-6.6%
superdense_keys	superdense_zipper_prev_step_iter/3200	23.01 µs	21.49 µs	-6.6%
sparse_keys	sparse_meet/50	888 ns	946 ns	+6.5%
shakespeare	shakespeare/shakespeare_sentences_val_count	2.85 ms	2.67 ms	-6.4%
superdense_keys	superdense_drop_bench/100	53 ns	50 ns	-6.3%
sparse_keys	sparse_iter/10	286 ns	269 ns	-6.0%
act_paths	size_paths_to_act/400000	58.50 ms	55.08 ms	-5.8%
zipper_head_owned	zipper_head_owned/map_reader_get_focus	30.05 µs	31.78 µs	+5.7%
sparse_keys	sparse_meet/1600	39.16 µs	41.26 µs	+5.4%
sparse_keys	sparse_meet/100	1.85 µs	1.94 µs	+5.1%

Full tables per bench are in the job log and the bench-out artifact.