blyxxyz:clean-shuf-2 - Branch - uutils/coreutils

shuf: Add `--random-seed`, make `--random-source` GNU-compatible, report write failures, optimize

#7585

Comparing

blyxxyz:clean-shuf-2

(

b4b40a6

) with

main

(

2d718f2

)

+84%

Improvements: 3

Untouched: 124

Skipped: 6

Benchmarks

Skipped (6)

Improved

shuf_input_range[1000000]

src/uu/shuf/benches/shuf_bench.rs

+84%

229.4 ms124.9 ms

shuf_lines[100000]

src/uu/shuf/benches/shuf_bench.rs

+17%

31.6 ms27.1 ms

shuf_repeat_sampling[50000]

src/uu/shuf/benches/shuf_bench.rs

+8%

5.1 ms4.7 ms

Passed

du_summarize_balanced_tree[(5, 4, 10)]

src/uu/du/benches/du_bench.rs

+2%

8.4 ms8.3 ms

tsort_input_parsing_heavy[5000]

src/uu/tsort/benches/tsort_bench.rs

+1%

85.9 ms84.9 ms

du_max_depth_balanced_tree[(6, 4, 10)]

src/uu/du/benches/du_bench.rs

+1%

32.7 ms32.4 ms

du_deep_tree[(100, 3)]

src/uu/du/benches/du_bench.rs

+1%

1.2 ms1.2 ms

cp_large_file[16]

src/uu/cp/benches/cp_bench.rs

362.9 µs361.5 µs

split_bytes

src/uu/split/benches/split_bench.rs

509.6 µs508.3 µs

split_number_chunks

src/uu/split/benches/split_bench.rs

294.7 µs293.9 µs

cp_recursive_balanced_tree[(5, 4, 10)]

src/uu/cp/benches/cp_bench.rs

72.4 ms72.2 ms

cp_recursive_wide_tree[(6000, 800)]

src/uu/cp/benches/cp_bench.rs

189.5 ms189.1 ms

cp_archive_balanced_tree[(5, 4, 10)]

src/uu/cp/benches/cp_bench.rs

78.1 ms77.9 ms

seq_formatted

src/uu/seq/benches/seq_bench.rs

96.6 ms96.4 ms

du_wide_tree[(5000, 500)]

src/uu/du/benches/du_bench.rs

11.3 ms11.3 ms

cp_recursive_deep_tree[(120, 4)]

src/uu/cp/benches/cp_bench.rs

13.6 ms13.6 ms

du_all_wide_tree[(5000, 500)]

src/uu/du/benches/du_bench.rs

18.1 ms18.1 ms

ls_recursive_long_all_deep_tree[(100, 4)]

src/uu/ls/benches/ls_bench.rs

3.3 ms3.3 ms

split_numeric_suffix

src/uu/split/benches/split_bench.rs

5.9 ms5.9 ms

cp_preserve_metadata[(5, 4, 10)]

src/uu/cp/benches/cp_bench.rs

74.9 ms74.8 ms

factor_multiple_u64s[2]

src/uu/factor/benches/factor_bench.rs

211.7 ms211.6 ms

sort_mixed_utf8_locale

src/uu/sort/benches/sort_locale_bench.rs

38.8 ms38.8 ms

ls_recursive_balanced_tree[(6, 4, 15)]

src/uu/ls/benches/ls_bench.rs

55.3 ms55.3 ms

b64_encode_synthetic

src/uu/base64/benches/base64_bench.rs

162 µs161.9 µs

ls_recursive_mixed_tree

src/uu/ls/benches/ls_bench.rs

1.4 ms1.4 ms

sort_unique_mixed

src/uu/sort/benches/sort_locale_bench.rs

39.3 ms39.3 ms

b64_decode_ignore_garbage_synthetic

src/uu/base64/benches/base64_bench.rs

163.5 µs163.5 µs

split_lines

src/uu/split/benches/split_bench.rs

5.7 ms5.7 ms

ls_recursive_long_all_mixed_tree

src/uu/ls/benches/ls_bench.rs

2.9 ms2.9 ms

b64_decode_synthetic

src/uu/base64/benches/base64_bench.rs

166.6 µs166.5 µs

ls_recursive_wide_tree[(10000, 1000)]

src/uu/ls/benches/ls_bench.rs

51.5 ms51.5 ms

mv_single_file

src/uu/mv/benches/mv_bench.rs

135.3 ms135.3 ms

ls_recursive_deep_tree[(200, 2)]

src/uu/ls/benches/ls_bench.rs

2.2 ms2.2 ms

mv_force_overwrite

src/uu/mv/benches/mv_bench.rs

144.1 ms144 ms

seq_large_integers

src/uu/seq/benches/seq_bench.rs

2.2 ms2.2 ms

wc_bytes_synthetic[500]

src/uu/wc/benches/wc_bench.rs

185.5 µs185.5 µs

sort_long_line[160000]

src/uu/sort/benches/sort_bench.rs

1.7 ms1.7 ms

sort_reverse_mixed

src/uu/sort/benches/sort_locale_bench.rs

39.2 ms39.1 ms

sort_mixed_c_locale

src/uu/sort/benches/sort_locale_bench.rs

38.3 ms38.3 ms

sort_german_locale

src/uu/sort/benches/sort_locale_bench.rs

39.3 ms39.3 ms

sort_numeric[500000]

src/uu/sort/benches/sort_bench.rs

1.2 s1.2 s

tsort_wide_dag[100000]

src/uu/tsort/benches/tsort_bench.rs

133.4 ms133.4 ms

hashsum_md5

src/uu/hashsum/benches/hashsum_bench.rs

32 ms32 ms

sort_case_insensitive[500000]

src/uu/sort/benches/sort_bench.rs

276.7 ms276.7 ms

rm_single_file

src/uu/rm/benches/rm_bench.rs

119.1 ms119.1 ms

seq_with_step

src/uu/seq/benches/seq_bench.rs

13.3 ms13.3 ms

hashsum_sha512

src/uu/hashsum/benches/hashsum_bench.rs

110.8 ms110.8 ms

seq_custom_separator

src/uu/seq/benches/seq_bench.rs

28.7 ms28.7 ms

seq_integers

src/uu/seq/benches/seq_bench.rs

26.1 ms26.1 ms

sort_accented_data[500000]

src/uu/sort/benches/sort_bench.rs

362.6 ms362.6 ms

cksum_sha3

src/uu/cksum/benches/cksum_bench.rs

1.8 s1.8 s

tsort_complex_dag[50000]

src/uu/tsort/benches/tsort_bench.rs

84 ms84 ms

wc_lines_large_line_count[500000]

src/uu/wc/benches/wc_bench.rs

2.9 ms2.9 ms

mv_directory

src/uu/mv/benches/mv_bench.rs

3.5 ms3.5 ms

sort_german_c_locale

src/uu/sort/benches/sort_locale_bench.rs

38.4 ms38.4 ms

tsort_tree_dag[(10, 3)]

src/uu/tsort/benches/tsort_bench.rs

34.1 ms34.1 ms

wc_chars_large_line_count[100000]

src/uu/wc/benches/wc_bench.rs

1 ms1 ms

numfmt_to_si[10000]

src/uu/numfmt/benches/numfmt_bench.rs

33.8 ms33.8 ms

numfmt_to_iec[10000]

src/uu/numfmt/benches/numfmt_bench.rs

33.8 ms33.8 ms

rm_force_files

src/uu/rm/benches/rm_bench.rs

2 ms2 ms

numfmt_round_modes[("up", 10000)]

src/uu/numfmt/benches/numfmt_bench.rs

34.3 ms34.3 ms

numfmt_round_modes[("towards-zero", 10000)]

src/uu/numfmt/benches/numfmt_bench.rs

34.3 ms34.3 ms

cut_fields_tab

src/uu/cut/benches/cut_bench.rs

34.1 ms34.1 ms

numfmt_from_si[10000]

src/uu/numfmt/benches/numfmt_bench.rs

32.7 ms32.7 ms

cksum_multiple_files

src/uu/cksum/benches/cksum_bench.rs

25.2 ms25.2 ms

sort_reverse_locale[500000]

src/uu/sort/benches/sort_bench.rs

361.8 ms361.8 ms

expand_many_lines[100000]

src/uu/expand/benches/expand_bench.rs

148.6 ms148.6 ms

nl_large_file[10]

src/uu/nl/benches/nl_bench.rs

57.1 ms57.1 ms

wc_default_large_line_count[100000]

src/uu/wc/benches/wc_bench.rs

50.4 ms50.4 ms

sort_ascii_only[500000]

src/uu/sort/benches/sort_bench.rs

353.8 ms353.8 ms

numfmt_large_numbers_si[10000]

src/uu/numfmt/benches/numfmt_bench.rs

32.5 ms32.5 ms

sort_mixed_data[500000]

src/uu/sort/benches/sort_bench.rs

328.1 ms328.1 ms

numfmt_round_modes[("down", 10000)]

src/uu/numfmt/benches/numfmt_bench.rs

34.3 ms34.3 ms

hashsum_sha1

src/uu/hashsum/benches/hashsum_bench.rs

72.5 ms72.5 ms

wc_words_large_line_count[100000]

src/uu/wc/benches/wc_bench.rs

48 ms48 ms

nl_many_lines[100000]

src/uu/nl/benches/nl_bench.rs

44.9 ms44.9 ms

unexpand_many_lines[100000]

src/uu/unexpand/benches/unexpand_bench.rs

269.6 ms269.6 ms

wc_bytes_lines_synthetic[2000]

src/uu/wc/benches/wc_bench.rs

120.2 ms120.2 ms

unexpand_large_file[10]

src/uu/unexpand/benches/unexpand_bench.rs

565.2 ms565.2 ms

cksum_shake256

src/uu/cksum/benches/cksum_bench.rs

1.9 s1.9 s

wc_words_synthetic[2000]

src/uu/wc/benches/wc_bench.rs

11.4 s11.4 s

cksum_shake128

src/uu/cksum/benches/cksum_bench.rs

1.5 s1.5 s

hashsum_sha256_check

src/uu/hashsum/benches/hashsum_bench.rs

172.9 ms172.9 ms

hashsum_md5_check

src/uu/hashsum/benches/hashsum_bench.rs

32 ms32 ms

fold_custom_width[50000]

src/uu/fold/benches/fold_bench.rs

33 ms33 ms

wc_lines_variable_length[(50, 500)]

src/uu/wc/benches/wc_bench.rs

3.4 ms3.4 ms

expand_custom_tabstops[50000]

src/uu/expand/benches/expand_bench.rs

36.9 ms36.9 ms

cut_fields_custom_delim

src/uu/cut/benches/cut_bench.rs

40.8 ms40.8 ms

fold_many_lines[100000]

src/uu/fold/benches/fold_bench.rs

80.4 ms80.4 ms

numfmt_to_si_precision[10000]

src/uu/numfmt/benches/numfmt_bench.rs

35.3 ms35.3 ms

sort_dictionary_order[500000]

src/uu/sort/benches/sort_bench.rs

1.1 s1.1 s

sort_numeric

src/uu/sort/benches/sort_locale_bench.rs

23.9 ms23.9 ms

rm_multiple_files

src/uu/rm/benches/rm_bench.rs

2.2 ms2.2 ms

sort_ascii_utf8_locale

src/uu/sort/benches/sort_locale_bench.rs

43.1 ms43.1 ms

sort_unique_locale[500000]

src/uu/sort/benches/sort_bench.rs

494.1 ms494.1 ms

cut_bytes

src/uu/cut/benches/cut_bench.rs

8.4 ms8.4 ms

sort_ascii_c_locale

src/uu/sort/benches/sort_locale_bench.rs

21.5 ms21.5 ms

hashsum_sha256

src/uu/hashsum/benches/hashsum_bench.rs

173 ms173 ms

cksum_blake2b

src/uu/cksum/benches/cksum_bench.rs

192.7 ms192.8 ms

cksum_sm3

src/uu/cksum/benches/cksum_bench.rs

1.6 s1.6 s

cksum_sha512

src/uu/cksum/benches/cksum_bench.rs

1.1 s1.1 s

cksum_sha384

src/uu/cksum/benches/cksum_bench.rs

1.1 s1.1 s

cut_characters

src/uu/cut/benches/cut_bench.rs

7.4 ms7.4 ms

uniq_case_insensitive[10000]

src/uu/uniq/benches/uniq_bench.rs

1.8 ms1.8 ms

numfmt_padding[(10000, 50)]

src/uu/numfmt/benches/numfmt_bench.rs

48 ms48 ms

wc_lines_extreme_line_lengths[(100000, 200)]

src/uu/wc/benches/wc_bench.rs

1.6 ms1.6 ms

cksum_sha1

src/uu/cksum/benches/cksum_bench.rs

723 ms723.1 ms

tsort_linear_chain[1000000]

src/uu/tsort/benches/tsort_bench.rs

1.5 s1.5 s

cksum_sha2

src/uu/cksum/benches/cksum_bench.rs

1.7 s1.7 s

uniq_with_count[10000]

src/uu/uniq/benches/uniq_bench.rs

1.9 ms1.9 ms

uniq_heavy_duplicates[10000]

src/uu/uniq/benches/uniq_bench.rs

2 ms2 ms

cksum_sha224

src/uu/cksum/benches/cksum_bench.rs

1.7 s1.7 s

cksum_sha256

src/uu/cksum/benches/cksum_bench.rs

1.7 s1.7 s

cksum_md5

src/uu/cksum/benches/cksum_bench.rs

317.9 ms317.9 ms

ls_recursive_long_all_balanced_tree[(6, 4, 15)]

src/uu/ls/benches/ls_bench.rs

134.5 ms134.5 ms

ls_recursive_long_all_wide_tree[(15000, 1500)]

src/uu/ls/benches/ls_bench.rs

143 ms143.1 ms

cksum_bsd

src/uu/cksum/benches/cksum_bench.rs

128.3 ms128.4 ms

cksum_blake3

src/uu/cksum/benches/cksum_bench.rs

208.9 µs209 µs

sort_key_field[500000]

src/uu/sort/benches/sort_bench.rs

701.3 ms701.7 ms

sort_case_sensitive[500000]

src/uu/sort/benches/sort_bench.rs

173.8 ms174 ms

rm_recursive_tree

src/uu/rm/benches/rm_bench.rs

11.7 ms11.8 ms

cksum_sysv

src/uu/cksum/benches/cksum_bench.rs

48.2 ms48.3 ms

mv_multiple_to_dir

src/uu/mv/benches/mv_bench.rs

12.9 ms12.9 ms

cksum_crc

src/uu/cksum/benches/cksum_bench.rs

16.9 ms17 ms

cksum_default

src/uu/cksum/benches/cksum_bench.rs

16.9 ms17 ms

cksum_crc32b

src/uu/cksum/benches/cksum_bench.rs

-1%

13.3 ms13.3 ms

cksum_raw_output

src/uu/cksum/benches/cksum_bench.rs

-1%

16.9 ms17 ms

Commits

Click on a commit to change the comparison range

Base

main

2d718f2

+83.69%

shuf: correctness: Use Fisher-Yates for nonrepeating integers We used to use a clever homegrown way to sample integers. But GNU shuf with --random-source observably uses Fisher-Yates, and the output of the old version depended on a heuristic (making it dangerous for --random-seed). So now we do Fisher-Yates here, just like we do for other inputs. In deterministic modes the output for --input-range is identical that for piping `seq` into `shuf`. We imitate the old algorithm's method for keeping the resource use in check. The performance of the new version is very close to that of the old version: I haven't found any cases where it's much faster or much slower.

cbc7c42

10 months ago

by blyxxyz

-0.02%

Merge branch 'main' into clean-shuf-2

b4b40a6

15 days ago

by sylvestre

Home Terms Privacy Docs