[connortsui20]

Tracking Issue: #10701

Top of the decimal Widen stack, on #11005. No release gate.

Summary

A same-scale precision increase copied every value into a wider buffer even when the caller did not need the values yet. Returning a Widen defers that copy and lets subsequent kernels keep working on the narrow child.

Changes

The child retains the source dtype and statistics because its values are unchanged. A cast of a decimal Widen passes to that child. Execution copies into the logical width directly, since yielding another widening cast would return a Widen again. The default-writer test compares round-trip values.

Known cost: the retained benchmarks show slower filtering at precision 38 (see #10946).

Historical local benchmark results

These retained measurements predate this style pass and rebase. They do not measure the current head.

cargo bench -p vortex-array --bench widen_decimal, Apple M4, medians. The numbers were measured at the top of this stack before the split, so they cover the whole stack, not this layer alone. The baseline is the canonical input, and the candidate is the same values as a Widen (child precision 5 under precision 10, 9 under 18, and 18 under 38). CodSpeed on this PR measures the layer itself.

bench	p10 canonical	p10 Widen	p18 canonical	p18 Widen	p38 canonical	p38 Widen
cast_widen	12.08 µs	6.958 µs	12.33 µs	6.958 µs	16.54 µs	14.58 µs

API Changes

A same-scale decimal cast returns a Widen when the target precision grows the canonical width.

[codspeed]

Merging this PR will regress 5 benchmarks

1 warning

⚠️ 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

⚡ 10 improved benchmarks
❌ 5 regressed benchmarks
✅ 2345 untouched benchmarks

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	CPU Simulation	bench_compare_varbin[(50000, 2)]	132 µs	162 µs	-18.49%
❌	CPU Simulation	bench_compare_varbin[(50000, 64)]	133.8 µs	163.6 µs	-18.23%
❌	CPU Simulation	bench_compare_sliced_dict_varbinview[(7500, 10000)]	199.6 µs	227.4 µs	-12.24%
❌	CPU Simulation	bench_compare_sliced_dict_varbinview[(10000, 10000)]	205.8 µs	233.6 µs	-11.93%
❌	CPU Simulation	bench_compare_sliced_dict_varbinview[(1000, 10000)]	77.9 µs	87.8 µs	-11.31%
⚡	CPU Simulation	cast_widen[p38_childp18_widen]	2.9 ms	1.5 ms	+95.67%
⚡	Walltime	mul_u64_nonnull_neon	39.1 µs	20.2 µs	+93.69%
⚡	CPU Simulation	widen_nullable[16384]	359 µs	224.9 µs	+59.59%
⚡	CPU Simulation	cast_widen[p10_childp5_widen]	1,211.4 µs	789.1 µs	+53.52%
⚡	CPU Simulation	cast_widen[p18_childp9_widen]	1,211.6 µs	789.5 µs	+53.47%
⚡	CPU Simulation	cast_widen[p38_childp18_canonical]	2.4 ms	1.8 ms	+34.16%
⚡	CPU Simulation	filter_powerlaw_by_random[10000]	1.6 µs	1.4 µs	+16.73%
⚡	CPU Simulation	take_filter_primitive_small_uncached_random_mask_random_indices[4096, 10]	23.8 µs	20.9 µs	+14.21%
⚡	CPU Simulation	cast_widen[p10_childp5_canonical]	1,009 µs	907.4 µs	+11.19%
⚡	CPU Simulation	cast_widen[p18_childp9_canonical]	1,008.5 µs	907 µs	+11.19%

Tip

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

---

_{Comparing ct/decimal-widen-lazy-cast (778cd7f) with ct/decimal-widen-stored (b3b006c)}