[DO NOT MERGE] Experimental `RowFn` by connortsui20 · Pull Request #9255 · vortex-data/vortex

Performance Regression: -8.37%

⚡ 33 improved benchmarks
❌ 50 regressed benchmarks
✅ 1879 untouched benchmarks
🆕 51 new benchmarks
⏩ 89 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

	Mode	Benchmark	`BASE`	`HEAD`	Efficiency
❌	Simulation	`case_when_nary_10_conditions[10000]`	750.1 µs	1,825 µs	-58.9%
❌	Simulation	`baseline_eq[4, 65536]`	586.5 µs	1,221.1 µs	-51.97%
❌	Simulation	`case_when_nary_equality_lookup[10000]`	517.8 µs	1,065.9 µs	-51.42%
❌	Simulation	`baseline_lt[4, 65536]`	602.2 µs	1,222.3 µs	-50.73%
❌	Simulation	`baseline_eq[16, 65536]`	690.6 µs	1,324.5 µs	-47.86%
❌	Simulation	`mul_i32_constant`	362.6 µs	689.4 µs	-47.4%
❌	Simulation	`baseline_lt[16, 65536]`	706.5 µs	1,325.3 µs	-46.69%
❌	Simulation	`case_when_nary_early_dominant[10000]`	385.3 µs	713.8 µs	-46.02%
❌	Simulation	`case_when_nary_3_conditions[10000]`	387.1 µs	716.3 µs	-45.96%
❌	Simulation	`case_when_all_false[100000]`	1.1 ms	2 ms	-44.65%
❌	Simulation	`case_when_all_true[100000]`	1.1 ms	2 ms	-44.47%
❌	Simulation	`case_when_simple[100000]`	1.2 ms	2.1 ms	-44.34%
❌	Simulation	`case_when_without_else[100000]`	1.3 ms	2.2 ms	-42.68%
❌	Simulation	`case_when_nary_10_conditions[1000]`	404.7 µs	690 µs	-41.34%
❌	Simulation	`case_when_all_false[10000]`	212.7 µs	325.5 µs	-34.66%
❌	Simulation	`case_when_all_true[10000]`	216.4 µs	330.2 µs	-34.46%
❌	Simulation	`sub_i64_constant`	636.5 µs	966.2 µs	-34.13%
❌	Simulation	`add_i64_constant`	636.4 µs	957.1 µs	-33.5%
❌	Simulation	`case_when_nary_equality_lookup[1000]`	311.8 µs	463.4 µs	-32.72%
❌	Simulation	`case_when_simple[10000]`	244.5 µs	356.2 µs	-31.36%
...	...	...	...	...	...

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 (0650a3f) with develop (402a5ac)}

89 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. ↩