CaseWhen uses forward pass with a remaining mask #6804

Performance Gate Passed

⚡ 10 improved benchmarks
✅ 384 untouched benchmarks
🆕 3 new benchmarks
⏩ 2052 skipped benchmarks¹

	Mode	Benchmark	`BASE`	`HEAD`	Efficiency
🆕	Simulation	`case_when_nary_early_dominant[10000]`	N/A	371 µs	N/A
🆕	Simulation	`case_when_nary_early_dominant[100000]`	N/A	2.3 ms	N/A
🆕	Simulation	`case_when_nary_early_dominant[1000]`	N/A	178.8 µs	N/A
⚡	Simulation	`case_when_nary_equality_lookup[100000]`	4.1 ms	3.2 ms	+29.68%
⚡	Simulation	`case_when_nary_equality_lookup[10000]`	622.7 µs	516.3 µs	+20.62%
⚡	Simulation	`case_when_nary_10_conditions[100000]`	7.3 ms	4.6 ms	+60.01%
⚡	Simulation	`case_when_nary_3_conditions[100000]`	2.9 ms	2.2 ms	+29.14%
⚡	Simulation	`case_when_nary_3_conditions[10000]`	432.1 µs	368 µs	+17.4%
⚡	Simulation	`case_when_nary_10_conditions[10000]`	1,047.2 µs	745.9 µs	+40.4%
⚡	Simulation	`case_when_nary_10_conditions[1000]`	400.6 µs	352.5 µs	+13.65%
⚡	Simulation	`take_map[(0.1, 0.1)]`	1,007.5 µs	905.4 µs	+11.28%
⚡	Simulation	`take_map[(0.1, 0.5)]`	2.6 ms	2.1 ms	+25.72%
⚡	Simulation	`take_map[(0.1, 1.0)]`	4.2 ms	3.5 ms	+18.36%

_{Comparing bp/case-when (cb59e4f) with develop (5d6a3c8)²}

2052 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. ↩
No successful run was found on develop (b1c5926) during the generation of this report, so 5d6a3c8 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩