naoNao89:feature/o-direct-buffer-alignment - Branch - uutils/coreutils

dd: optimize O_DIRECT buffer alignment to reduce syscall overhead

#9104

Comparing

naoNao89:feature/o-direct-buffer-alignment

(

114a50b

) with

main

(

677fd95

)

Untouched: 125

No successful run was found on main ( 8983b90) during the generation of this report, so 677fd95 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report.

Benchmarks

Passed

du_max_depth_balanced_tree[(6, 4, 10)]

src/uu/du/benches/du_bench.rs

+1%

33.3 ms33 ms

tsort_input_parsing_heavy[5000]

src/uu/tsort/benches/tsort_bench.rs

+1%

120.3 ms119.7 ms

sort_reverse_locale[500000]

src/uu/sort/benches/sort_bench.rs

+1%

363.9 ms362.1 ms

sort_ascii_only[500000]

src/uu/sort/benches/sort_bench.rs

355.4 ms353.6 ms

du_all_balanced_tree[(4, 3, 10)]

src/uu/du/benches/du_bench.rs

1.6 ms1.6 ms

du_wide_tree[(5000, 500)]

src/uu/du/benches/du_bench.rs

11.3 ms11.2 ms

cp_preserve_metadata[(5, 4, 10)]

src/uu/cp/benches/cp_bench.rs

74.9 ms74.8 ms

du_all_wide_tree[(5000, 500)]

src/uu/du/benches/du_bench.rs

18.2 ms18.2 ms

cp_large_file[16]

src/uu/cp/benches/cp_bench.rs

343.7 µs343.4 µs

cp_recursive_wide_tree[(6000, 800)]

src/uu/cp/benches/cp_bench.rs

189.1 ms189 ms

sort_german_c_locale

src/uu/sort/benches/sort_locale_bench.rs

38.4 ms38.4 ms

sort_german_locale

src/uu/sort/benches/sort_locale_bench.rs

39.1 ms39.1 ms

sort_numeric[500000]

src/uu/sort/benches/sort_bench.rs

1.2 s1.2 s

split_number_chunks

src/uu/split/benches/split_bench.rs

275.6 µs275.5 µs

cp_recursive_deep_tree[(120, 4)]

src/uu/cp/benches/cp_bench.rs

13.6 ms13.6 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

149.8 µs149.8 µs

sort_mixed_utf8_locale

src/uu/sort/benches/sort_locale_bench.rs

38.8 ms38.8 ms

b64_decode_synthetic

src/uu/base64/benches/base64_bench.rs

149.2 µs149.2 µs

tsort_complex_dag[50000]

src/uu/tsort/benches/tsort_bench.rs

84.6 ms84.6 ms

ls_recursive_long_all_mixed_tree

src/uu/ls/benches/ls_bench.rs

2.8 ms2.8 ms

du_summarize_balanced_tree[(5, 4, 10)]

src/uu/du/benches/du_bench.rs

8.4 ms8.4 ms

tsort_tree_dag[(10, 3)]

src/uu/tsort/benches/tsort_bench.rs

34.4 ms34.4 ms

ls_recursive_long_all_deep_tree[(100, 4)]

src/uu/ls/benches/ls_bench.rs

3.2 ms3.2 ms

sort_long_line[160000]

src/uu/sort/benches/sort_bench.rs

1.6 ms1.6 ms

mv_force_overwrite

src/uu/mv/benches/mv_bench.rs

125.1 ms125.1 ms

rm_single_file

src/uu/rm/benches/rm_bench.rs

106.5 ms106.5 ms

cp_archive_balanced_tree[(5, 4, 10)]

src/uu/cp/benches/cp_bench.rs

77.8 ms77.8 ms

split_numeric_suffix

src/uu/split/benches/split_bench.rs

5.9 ms5.9 ms

uniq_heavy_duplicates[10000]

src/uu/uniq/benches/uniq_bench.rs

1.9 ms1.9 ms

mv_single_file

src/uu/mv/benches/mv_bench.rs

129.6 ms129.6 ms

wc_lines_extreme_line_lengths[(100000, 200)]

src/uu/wc/benches/wc_bench.rs

1.6 ms1.6 ms

tsort_linear_chain[1000000]

src/uu/tsort/benches/tsort_bench.rs

1.5 s1.5 s

sort_case_insensitive[500000]

src/uu/sort/benches/sort_bench.rs

278.6 ms278.5 ms

sort_numeric

src/uu/sort/benches/sort_locale_bench.rs

23.2 ms23.2 ms

sort_reverse_mixed

src/uu/sort/benches/sort_locale_bench.rs

39.1 ms39.1 ms

wc_bytes_synthetic[500]

src/uu/wc/benches/wc_bench.rs

164.2 µs164.2 µs

wc_lines_large_line_count[500000]

src/uu/wc/benches/wc_bench.rs

2.9 ms2.9 ms

cut_bytes

src/uu/cut/benches/cut_bench.rs

8.5 ms8.5 ms

wc_lines_variable_length[(50, 500)]

src/uu/wc/benches/wc_bench.rs

3.4 ms3.4 ms

cut_characters

src/uu/cut/benches/cut_bench.rs

7.5 ms7.5 ms

rm_force_files

src/uu/rm/benches/rm_bench.rs

2.1 ms2.1 ms

ls_recursive_long_all_wide_tree[(15000, 1500)]

src/uu/ls/benches/ls_bench.rs

142.4 ms142.4 ms

mv_directory

src/uu/mv/benches/mv_bench.rs

3.5 ms3.5 ms

ls_recursive_mixed_tree

src/uu/ls/benches/ls_bench.rs

1.4 ms1.4 ms

ls_recursive_wide_tree[(10000, 1000)]

src/uu/ls/benches/ls_bench.rs

53 ms53 ms

uniq_case_insensitive[10000]

src/uu/uniq/benches/uniq_bench.rs

1.7 ms1.7 ms

numfmt_padding[(10000, 50)]

src/uu/numfmt/benches/numfmt_bench.rs

47.8 ms47.8 ms

wc_chars_large_line_count[100000]

src/uu/wc/benches/wc_bench.rs

994.5 µs994.5 µs

cut_fields_tab

src/uu/cut/benches/cut_bench.rs

34.1 ms34.1 ms

hashsum_md5_check

src/uu/hashsum/benches/hashsum_bench.rs

32 ms32 ms

cut_fields_custom_delim

src/uu/cut/benches/cut_bench.rs

40.7 ms40.7 ms

cksum_default

src/uu/cksum/benches/cksum_bench.rs

17.7 ms17.7 ms

numfmt_to_si_precision[10000]

src/uu/numfmt/benches/numfmt_bench.rs

35.5 ms35.5 ms

expand_custom_tabstops[50000]

src/uu/expand/benches/expand_bench.rs

36.6 ms36.6 ms

seq_custom_separator

src/uu/seq/benches/seq_bench.rs

28.7 ms28.7 ms

wc_bytes_lines_synthetic[2000]

src/uu/wc/benches/wc_bench.rs

120.2 ms120.2 ms

numfmt_to_iec[10000]

src/uu/numfmt/benches/numfmt_bench.rs

34.2 ms34.2 ms

expand_many_lines[100000]

src/uu/expand/benches/expand_bench.rs

148.1 ms148.1 ms

numfmt_from_si[10000]

src/uu/numfmt/benches/numfmt_bench.rs

33.4 ms33.4 ms

numfmt_large_numbers_si[10000]

src/uu/numfmt/benches/numfmt_bench.rs

32.9 ms32.9 ms

wc_default_large_line_count[100000]

src/uu/wc/benches/wc_bench.rs

50.4 ms50.4 ms

wc_words_synthetic[2000]

src/uu/wc/benches/wc_bench.rs

11.4 s11.4 s

cksum_shake256

src/uu/cksum/benches/cksum_bench.rs

1.9 s1.9 s

numfmt_to_si[10000]

src/uu/numfmt/benches/numfmt_bench.rs

34.2 ms34.2 ms

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

cksum_shake128

src/uu/cksum/benches/cksum_bench.rs

1.5 s1.5 s

cksum_sha224

src/uu/cksum/benches/cksum_bench.rs

1.7 s1.7 s

unexpand_large_file[10]

src/uu/unexpand/benches/unexpand_bench.rs

549.3 ms549.3 ms

cksum_sm3

src/uu/cksum/benches/cksum_bench.rs

1.6 s1.6 s

cksum_sha256

src/uu/cksum/benches/cksum_bench.rs

1.7 s1.7 s

cksum_sha2

src/uu/cksum/benches/cksum_bench.rs

1.7 s1.7 s

cksum_sha3

src/uu/cksum/benches/cksum_bench.rs

1.8 s1.8 s

nl_many_lines[100000]

src/uu/nl/benches/nl_bench.rs

79.8 ms79.8 ms

cksum_sha1

src/uu/cksum/benches/cksum_bench.rs

722.9 ms722.9 ms

cksum_md5

src/uu/cksum/benches/cksum_bench.rs

317.8 ms317.8 ms

hashsum_sha256_check

src/uu/hashsum/benches/hashsum_bench.rs

173.6 ms173.6 ms

cksum_blake2b

src/uu/cksum/benches/cksum_bench.rs

192.9 ms192.9 ms

numfmt_round_modes[("down", 10000)]

src/uu/numfmt/benches/numfmt_bench.rs

34.6 ms34.6 ms

unexpand_many_lines[100000]

src/uu/unexpand/benches/unexpand_bench.rs

262 ms262 ms

sort_dictionary_order[500000]

src/uu/sort/benches/sort_bench.rs

1.1 s1.1 s

nl_large_file[10]

src/uu/nl/benches/nl_bench.rs

100.9 ms100.9 ms

numfmt_round_modes[("up", 10000)]

src/uu/numfmt/benches/numfmt_bench.rs

34.6 ms34.6 ms

rm_recursive_tree

src/uu/rm/benches/rm_bench.rs

13.6 ms13.6 ms

wc_words_large_line_count[100000]

src/uu/wc/benches/wc_bench.rs

47.9 ms47.9 ms

cksum_raw_output

src/uu/cksum/benches/cksum_bench.rs

17.8 ms17.8 ms

cksum_bsd

src/uu/cksum/benches/cksum_bench.rs

128.1 ms128.1 ms

cksum_crc

src/uu/cksum/benches/cksum_bench.rs

17.8 ms17.8 ms

seq_formatted

src/uu/seq/benches/seq_bench.rs

97.2 ms97.2 ms

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

src/uu/numfmt/benches/numfmt_bench.rs

34.7 ms34.7 ms

cksum_sysv

src/uu/cksum/benches/cksum_bench.rs

48 ms48 ms

fold_many_lines[100000]

src/uu/fold/benches/fold_bench.rs

116.2 ms116.2 ms

sort_key_field[500000]

src/uu/sort/benches/sort_bench.rs

700.1 ms700.1 ms

rm_multiple_files

src/uu/rm/benches/rm_bench.rs

2.3 ms2.3 ms

cksum_crc32b

src/uu/cksum/benches/cksum_bench.rs

14.1 ms14.1 ms

sort_accented_data[500000]

src/uu/sort/benches/sort_bench.rs

362.4 ms362.4 ms

cksum_multiple_files

src/uu/cksum/benches/cksum_bench.rs

26.6 ms26.6 ms

seq_integers

src/uu/seq/benches/seq_bench.rs

26.1 ms26.1 ms

fold_custom_width[50000]

src/uu/fold/benches/fold_bench.rs

43.4 ms43.4 ms

sort_case_sensitive[500000]

src/uu/sort/benches/sort_bench.rs

174.2 ms174.2 ms

uniq_with_count[10000]

src/uu/uniq/benches/uniq_bench.rs

1.9 ms1.9 ms

split_lines

src/uu/split/benches/split_bench.rs

5.7 ms5.7 ms

seq_with_step

src/uu/seq/benches/seq_bench.rs

13.3 ms13.3 ms

sort_ascii_utf8_locale

src/uu/sort/benches/sort_locale_bench.rs

43 ms43.1 ms

cp_recursive_balanced_tree[(5, 4, 10)]

src/uu/cp/benches/cp_bench.rs

72.2 ms72.2 ms

tsort_wide_dag[100000]

src/uu/tsort/benches/tsort_bench.rs

134.6 ms134.6 ms

ls_recursive_balanced_tree[(6, 4, 15)]

src/uu/ls/benches/ls_bench.rs

56.4 ms56.4 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

173.6 ms173.6 ms

hashsum_sha1

src/uu/hashsum/benches/hashsum_bench.rs

72.5 ms72.5 ms

ls_recursive_long_all_balanced_tree[(6, 4, 15)]

src/uu/ls/benches/ls_bench.rs

136.3 ms136.3 ms

hashsum_sha512

src/uu/hashsum/benches/hashsum_bench.rs

110.8 ms110.8 ms

ls_recursive_deep_tree[(200, 2)]

src/uu/ls/benches/ls_bench.rs

2.2 ms2.2 ms

hashsum_md5

src/uu/hashsum/benches/hashsum_bench.rs

32 ms32 ms

sort_unique_locale[500000]

src/uu/sort/benches/sort_bench.rs

494 ms494.1 ms

b64_encode_synthetic

src/uu/base64/benches/base64_bench.rs

146 µs146.1 µs

sort_mixed_data[500000]

src/uu/sort/benches/sort_bench.rs

327.3 ms327.3 ms

du_deep_tree[(100, 3)]

src/uu/du/benches/du_bench.rs

1.2 ms1.2 ms

sort_mixed_c_locale

src/uu/sort/benches/sort_locale_bench.rs

38.3 ms38.3 ms

mv_multiple_to_dir

src/uu/mv/benches/mv_bench.rs

13.1 ms13.1 ms

cksum_blake3

src/uu/cksum/benches/cksum_bench.rs

95.7 ms95.7 ms

du_human_balanced_tree[(5, 4, 10)]

src/uu/du/benches/du_bench.rs

10.5 ms10.5 ms

split_bytes

src/uu/split/benches/split_bench.rs

505.4 µs505.7 µs

du_balanced_tree[(5, 4, 10)]

src/uu/du/benches/du_bench.rs

9.2 ms9.3 ms

Commits

Click on a commit to change the comparison range

Base

main

677fd95

-0.01%

fix(dd): optimize O_DIRECT buffer alignment to reduce syscall overhead Implement page-aligned buffer allocation and optimize O_DIRECT flag handling to match GNU dd behavior. Key changes: - Add allocate_aligned_buffer() for page-aligned memory allocation - Update buffer allocation to use aligned buffers - Modify handle_o_direct_write() to only remove O_DIRECT for partial blocks - Add Output::write_with_o_direct_handling() for proper O_DIRECT handling - Add comprehensive unit and integration tests Fixes #6078

2560240

2 months ago

by naoNao89

fix(dd): fix test_o_direct_with_conv_flags to use compatible conv=sync flag

2719973

2 months ago

by naoNao89

-0.01%

fix(dd): remove test_o_direct_with_conv_flags - incompatible with O_DIRECT on ARM O_DIRECT requires page-aligned buffers and writes. The conv=sync flag pads output to block size, which may not be page-aligned, causing EINVAL errors on ARM systems. The core O_DIRECT functionality is already well-tested by: - test_o_direct_with_aligned_buffer_full_blocks - test_o_direct_with_partial_final_block - test_o_direct_various_block_sizes

a9b075b

2 months ago

by naoNao89

+0.03%

fix: remove trailing blank line in test_dd.rs to fix cargo fmt

114a50b

2 months ago

by naoNao89

Home Terms Privacy Docs