benchmarks: reduce allocations in run_util_function

[xtqqczze]

No description provided.

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/tail/follow-name (fails in this run but passes in the 'main' branch)
Congrats! The gnu test tests/dd/stderr is no longer failing!
Congrats! The gnu test tests/tac/tac-2-nonseekable is no longer failing!
Congrats! The gnu test tests/tail/follow-stdin is no longer failing!

[codspeed-hq]

CodSpeed Performance Report

Merging this PR will degrade performance by 29.9%

_{Comparing xtqqczze:allocs-run_util_function (66aa24d) with main (b302bbe)}

Summary

⚡ 9 improved benchmarks
❌ 15 regressed benchmarks
✅ 258 untouched benchmarks
⏩ 38 skipped benchmarks¹

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

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	Memory	sort_mixed_data[500000]	25.8 MB	21.9 MB	+17.83%
⚡	Memory	sort_key_field[500000]	32.8 MB	28.9 MB	+13.46%
⚡	Memory	sort_ascii_only[500000]	25.5 MB	21.6 MB	+17.98%
⚡	Memory	sort_accented_data[500000]	25.5 MB	21.6 MB	+18.1%
⚡	Memory	sort_numeric[500000]	48.7 MB	44.8 MB	+8.74%
❌	Memory	mv_single_file	12.1 MB	16.5 MB	-27.06%
❌	Memory	mv_force_overwrite	11.7 MB	16.7 MB	-29.9%
❌	Memory	seq_large_integers	80.2 KB	84.7 KB	-5.32%
❌	Memory	cksum_sha2	66.1 KB	70.7 KB	-6.63%
❌	Memory	cksum_sha512	55.7 KB	59.2 KB	-5.83%
❌	Memory	cksum_crc32b	54 KB	56.7 KB	-4.64%
❌	Memory	cksum_sha256	54.1 KB	57.3 KB	-5.57%
⚡	Memory	cksum_sysv	60.8 KB	55.7 KB	+9.21%
❌	Memory	cksum_sha1	61.8 KB	66.5 KB	-7.11%
❌	Memory	cksum_bsd	61.6 KB	63.8 KB	-3.49%
⚡	Memory	cksum_sha3	65.5 KB	60.1 KB	+8.88%
⚡	Memory	cksum_default	61.1 KB	58.5 KB	+4.33%
❌	Memory	cksum_md5	61.8 KB	64 KB	-3.48%
❌	Memory	cksum_crc	61.9 KB	66.6 KB	-7.04%
❌	Memory	cksum_sm3	61.8 KB	66.5 KB	-7.11%
...	...	...	...	...	...

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

Footnotes

1. 38 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. ↩

[xtqqczze]

Hang on, why are we constructing a Vec<std::ffi::OsString> inside the benchmarks themselves? That adds allocation and conversion overhead to the measured section and will skew the results.

[github-actions]

GNU testsuite comparison:

Skip an intermittent issue tests/tail/follow-name (fails in this run but passes in the 'main' branch)
Congrats! The gnu test tests/dd/stderr is no longer failing!
Congrats! The gnu test tests/tac/tac-2-nonseekable is no longer failing!
Congrats! The gnu test tests/tail/follow-stdin is no longer failing!

[xtqqczze]

Hang on, why are we constructing a Vec<std::ffi::OsString> inside the benchmarks themselves? That adds allocation and conversion overhead to the measured section and will skew the results.

#10388