mimalloc benchmark with Rust 1.95

[oech3]

No description provided.

[codspeed-hq]

Merging this PR will degrade performance by 26.46%

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

⚡ 68 improved benchmarks
❌ 19 regressed benchmarks
✅ 67 untouched benchmarks
⏩ 201 skipped benchmarks¹

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

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	Simulation	wc_bytes_synthetic[500]	81.6 µs	73.9 µs	+10.4%
⚡	Simulation	wc_bytes_synthetic[1]	83.8 µs	74.6 µs	+12.36%
❌	Simulation	sort_numeric_utf8_locale	20.2 ms	21 ms	-3.61%
❌	Simulation	sort_reverse_utf8_locale	35.5 ms	37 ms	-3.93%
❌	Simulation	sort_mixed_c_locale	35.6 ms	37.1 ms	-3.88%
❌	Simulation	sort_unique_utf8_locale	35.8 ms	37.3 ms	-3.99%
❌	Simulation	sort_ascii_utf8_locale	15.4 ms	17.5 ms	-12.18%
❌	Simulation	sort_mixed_utf8_locale	35.6 ms	37.1 ms	-4.04%
⚡	Simulation	sort_general_numeric[200000]	701.6 ms	658.4 ms	+6.56%
⚡	Simulation	sort_key_field[500000]	770.3 ms	742.3 ms	+3.76%
❌	Simulation	sort_long_line[10000]	424.7 µs	488.4 µs	-13.04%
❌	Simulation	sort_ascii_c_locale	16 ms	17.4 ms	-8.45%
⚡	Simulation	b64_decode_ignore_garbage_synthetic	63.3 µs	57.4 µs	+10.26%
⚡	Simulation	b64_decode_synthetic	63.5 µs	56.6 µs	+12.11%
⚡	Simulation	b64_encode_synthetic	59.2 µs	55 µs	+7.71%
⚡	Simulation	ls_recursive_deep_tree[(200, 2)]	1.5 ms	1.2 ms	+25.43%
⚡	Simulation	ls_recursive_long_all_wide_tree[(15000, 1500)]	88.2 ms	81.5 ms	+8.19%
⚡	Simulation	ls_recursive_wide_tree[(10000, 1000)]	32.9 ms	25.7 ms	+28.17%
⚡	Simulation	ls_recursive_long_all_balanced_tree[(6, 4, 15)]	109.8 ms	88.8 ms	+23.66%
⚡	Simulation	ls_recursive_long_all_deep_tree[(100, 4)]	2.5 ms	2.1 ms	+17.79%
...	...	...	...	...	...

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

---

_{Comparing oech3:nix-ctrlc (b02b26a) with main (4704cae)}

Footnotes

1. 201 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. ↩

[oech3]

I am confused. Why are true/false improved?

[xtqqczze]

I am confused. Why are true/false improved?

Different runtime environment: Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz → AMD EPYC 7763 64-Core Processor

[oech3]

Ofcause. But I saw similar result many times previously.

[oech3]

#11902 might reduce perf difference between default allocator.

[xtqqczze]

@oech3 could you please run again to try get a result from same runtime environment

[oech3]

I don't think we can 100% choice same runtime env.

[xtqqczze]

I don't think we can 100% choice same runtime env.

It's a known issue with benchmarks involving allocation: https://codspeed.io/blog/why-glibc-faster-github-actions