Add benchmark filtering in instrumentation mode

[GuillaumeLagrange]

No description provided.

[codspeed-hq]

CodSpeed Instrumentation Performance Report

Merging #127 will degrade performances by 24.48%

_{Comparing instrumentation-filter (16f9dfb) with main (9ace864)¹}

Summary

⚡ 13 improvements
❌ 12 regressions
✅ 161 untouched

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Benchmarks breakdown

	Benchmark	BASE	HEAD	Change
❌	sleep_100ms	536.9 ns	595.3 ns	-9.8%
❌	sleep_10ms	536.9 ns	595.3 ns	-9.8%
❌	sleep_1ms	536.9 ns	595.3 ns	-9.8%
❌	sleep_50ms	536.9 ns	595.3 ns	-9.8%
⚡	bench_array1[10]	225 ns	195.8 ns	+14.89%
⚡	bench_array2[10]	225 ns	195.8 ns	+14.89%
❌	fibo_10	90 ns	119.2 ns	-24.48%
⚡	fibo_50	119.4 ns	90.3 ns	+32.31%
❌	print_env_hello	7.8 µs	8.1 µs	-3.61%
⚡	iterative[0]	148.3 ns	119.2 ns	+24.48%
⚡	recursive[0]	240.3 ns	211.1 ns	+13.82%
⚡	recursive[5]	320.3 ns	291.1 ns	+10.02%
❌	recursive_memoized[BTreeMap<u64, u64>, 0]	371.7 ns	430 ns	-13.57%
⚡	recursive_memoized[HashMap<u64, u64>, 30]	13.3 µs	12.8 µs	+4.09%
❌	generate_parentheses[4]	16.7 µs	17.8 µs	-5.97%
⚡	add_two_integers[(0, 0)]	150.3 ns	121.1 ns	+24.08%
❌	find_highest_set_bit[1024]	91.9 ns	121.1 ns	-24.08%
❌	find_highest_set_bit[255]	91.9 ns	121.1 ns	-24.08%
❌	find_highest_set_bit[42]	91.9 ns	121.1 ns	-24.08%
❌	find_highest_set_bit[65535]	91.9 ns	121.1 ns	-24.08%
...	...	...	...	...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

Footnotes

1. No successful run was found on main (16f9dfb) during the generation of this report, so 9ace864 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

[codspeed-hq]

CodSpeed WallTime Performance Report

Merging #127 will degrade performances by 50%

_{Comparing instrumentation-filter (16f9dfb) with main (9ace864)}

Summary

⚡ 10 improvements
❌ 24 regressions
✅ 134 untouched

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Benchmarks breakdown

	Benchmark	BASE	HEAD	Change
❌	Iterative	20 ns	21 ns	-4.76%
❌	Recursive	50.6 µs	52.6 µs	-3.66%
❌	Iterative[20]	20 ns	21 ns	-4.76%
❌	Iterative[21]	21 ns	22 ns	-4.55%
❌	Recursive[20]	50.4 µs	52.3 µs	-3.78%
❌	Recursive[21]	82.5 µs	85.1 µs	-3.01%
❌	large_setup	18 ns	20 ns	-10%
❌	iter_with_setup	36 ns	44 ns	-18.18%
❌	iter_batched_large_input	40 ns	44 ns	-9.09%
❌	iter_batched_per_iteration	36 ns	44 ns	-18.18%
❌	iter_batched_ref_large_input	36 ns	44 ns	-18.18%
❌	iter_batched_ref_per_iteration	36 ns	44 ns	-18.18%
❌	iter_with_setup	36 ns	48 ns	-25%
❌	from_elem[2048]	255 ns	265 ns	-3.77%
❌	from_elem[4096]	340 ns	352 ns	-3.41%
⚡	from_elem_decimal[2048]	270 ns	252 ns	+7.14%
⚡	mut_borrow	8 ns	7 ns	+14.29%
❌	recursive[0]	1 ns	2 ns	-50%
⚡	recursive_memoized[BTreeMap<u64, u64>, 0]	18 ns	14 ns	+28.57%
❌	recursive_memoized[BTreeMap<u64, u64>, 5]	209 ns	221 ns	-5.43%
...	...	...	...	...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

[art049]

We need some tests if we want to release this confidently. There is a significant amount of changes in the compat layers and we also have already the tests for the walltime equivalent afaik