mirror of
https://github.com/syncthing/syncthing.git
synced 2026-10-26 04:58:28 -04:00
### Purpose Resurrecting https://github.com/syncthing/syncthing/pull/9365 ### Testing Current benchmark results: ``` goos: linux goarch: amd64 pkg: github.com/syncthing/syncthing/lib/fs cpu: AMD EPYC 7763 64-Core Processor │ ../old.txt │ ../new.txt │ │ sec/op │ sec/op vs base │ UnicodeLowercase/ASCII_lowercase-4 43.68n ± 4% 12.19n ± 1% -72.09% (p=0.000 n=10) UnicodeLowercase/ASCII_mixedcase_start-4 200.65n ± 2% 59.49n ± 3% -70.35% (p=0.000 n=10) UnicodeLowercase/ASCII_mixedcase_end-4 95.50n ± 2% 59.10n ± 2% -38.12% (p=0.000 n=10) UnicodeLowercase/Latin1_lowercase-4 122.5n ± 1% 131.4n ± 1% +7.27% (p=0.000 n=10) UnicodeLowercase/Latin1_mixedcase_start-4 339.9n ± 2% 309.2n ± 1% -9.05% (p=0.000 n=10) UnicodeLowercase/Latin1_mixedcase_end-4 183.6n ± 2% 174.3n ± 1% -5.04% (p=0.000 n=10) UnicodeLowercase/Unicode_lowercase-4 456.6n ± 1% 440.5n ± 1% -3.53% (p=0.000 n=10) UnicodeLowercase/Unicode_mixedcase_start-4 625.9n ± 1% 595.6n ± 1% -4.83% (p=0.000 n=10) UnicodeLowercase/Unicode_mixedcase_end-4 516.2n ± 1% 495.5n ± 1% -4.02% (p=0.000 n=10) geomean 214.1n 150.4n -29.72% │ ../old.txt │ ../new.txt │ │ B/op │ B/op vs base │ UnicodeLowercase/ASCII_lowercase-4 0.000 ± 0% 0.000 ± 0% ~ (p=1.000 n=10) ¹ UnicodeLowercase/ASCII_mixedcase_start-4 24.00 ± 0% 24.00 ± 0% ~ (p=1.000 n=10) ¹ UnicodeLowercase/ASCII_mixedcase_end-4 24.00 ± 0% 24.00 ± 0% ~ (p=1.000 n=10) ¹ UnicodeLowercase/Latin1_lowercase-4 0.000 ± 0% 0.000 ± 0% ~ (p=1.000 n=10) ¹ UnicodeLowercase/Latin1_mixedcase_start-4 32.00 ± 0% 32.00 ± 0% ~ (p=1.000 n=10) ¹ UnicodeLowercase/Latin1_mixedcase_end-4 32.00 ± 0% 32.00 ± 0% ~ (p=1.000 n=10) ¹ UnicodeLowercase/Unicode_lowercase-4 0.000 ± 0% 0.000 ± 0% ~ (p=1.000 n=10) ¹ UnicodeLowercase/Unicode_mixedcase_start-4 48.00 ± 0% 48.00 ± 0% ~ (p=1.000 n=10) ¹ UnicodeLowercase/Unicode_mixedcase_end-4 48.00 ± 0% 48.00 ± 0% ~ (p=1.000 n=10) ¹ geomean ² +0.00% ² ¹ all samples are equal ² summaries must be >0 to compute geomean │ ../old.txt │ ../new.txt │ │ allocs/op │ allocs/op vs base │ UnicodeLowercase/ASCII_lowercase-4 0.000 ± 0% 0.000 ± 0% ~ (p=1.000 n=10) ¹ UnicodeLowercase/ASCII_mixedcase_start-4 1.000 ± 0% 1.000 ± 0% ~ (p=1.000 n=10) ¹ UnicodeLowercase/ASCII_mixedcase_end-4 1.000 ± 0% 1.000 ± 0% ~ (p=1.000 n=10) ¹ UnicodeLowercase/Latin1_lowercase-4 0.000 ± 0% 0.000 ± 0% ~ (p=1.000 n=10) ¹ UnicodeLowercase/Latin1_mixedcase_start-4 1.000 ± 0% 1.000 ± 0% ~ (p=1.000 n=10) ¹ UnicodeLowercase/Latin1_mixedcase_end-4 1.000 ± 0% 1.000 ± 0% ~ (p=1.000 n=10) ¹ UnicodeLowercase/Unicode_lowercase-4 0.000 ± 0% 0.000 ± 0% ~ (p=1.000 n=10) ¹ UnicodeLowercase/Unicode_mixedcase_start-4 1.000 ± 0% 1.000 ± 0% ~ (p=1.000 n=10) ¹ UnicodeLowercase/Unicode_mixedcase_end-4 1.000 ± 0% 1.000 ± 0% ~ (p=1.000 n=10) ¹ geomean ² +0.00% ² ¹ all samples are equal ² summaries must be >0 to compute geomean ``` I think the `+7%` for the lowercase Latin1 testcase is easily outweighed by the ASCII and unicode improvements 🙂