feat(dd): add first benchmark suite for performance validation #9136
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
- Create dd's first benchmark suite using divan framework - Benchmark various block sizes (4K, 8K, 64K, 1M) to measure performance - Test different dd scenarios: default, partial copy, skip, seek operations - Measure impact of separate input/output block sizes - All benchmarks use status=none to avoid output noise - Benchmarks verify the O_DIRECT buffer alignment optimization - Follows existing uutils benchmark patterns and conventions
CodSpeed Performance ReportMerging #9136 will not alter performanceComparing Summary
Benchmarks breakdown
Footnotes
|
|
GNU testsuite comparison: |
|
GNU testsuite comparison: |
Contributor
|
can we have tests that take between 100 to 300ms? |
Increase benchmark dataset sizes to achieve consistent 100-300ms timing: - dd_copy_default: 16 -> 32 MB - dd_copy_4k_blocks: 16 -> 24 MB - dd_copy_64k_blocks: 16 -> 64 MB - dd_copy_1m_blocks: 16 -> 128 MB - dd_copy_separate_blocks: 16 -> 48 MB - dd_copy_partial: 16 -> 32 MB - dd_copy_with_skip: 16 -> 48 MB - dd_copy_with_seek: 16 -> 48 MB - dd_copy_8k_blocks: 16 -> 32 MB This ensures stable, repeatable benchmark measurements across different systems.
|
GNU testsuite comparison: |
naoNao89
added a commit
to naoNao89/coreutils
that referenced
this pull request
Dec 31, 2025
…s#9136) * feat(dd): add comprehensive benchmark suite for O_DIRECT optimization - Create dd's first benchmark suite using divan framework - Benchmark various block sizes (4K, 8K, 64K, 1M) to measure performance - Test different dd scenarios: default, partial copy, skip, seek operations - Measure impact of separate input/output block sizes - All benchmarks use status=none to avoid output noise - Benchmarks verify the O_DIRECT buffer alignment optimization - Follows existing uutils benchmark patterns and conventions * bench(dd): increase dataset sizes for consistent timing Increase benchmark dataset sizes to achieve consistent 100-300ms timing: - dd_copy_default: 16 -> 32 MB - dd_copy_4k_blocks: 16 -> 24 MB - dd_copy_64k_blocks: 16 -> 64 MB - dd_copy_1m_blocks: 16 -> 128 MB - dd_copy_separate_blocks: 16 -> 48 MB - dd_copy_partial: 16 -> 32 MB - dd_copy_with_skip: 16 -> 48 MB - dd_copy_with_seek: 16 -> 48 MB - dd_copy_8k_blocks: 16 -> 32 MB This ensures stable, repeatable benchmark measurements across different systems. --------- Co-authored-by: Sylvestre Ledru <sylvestre@debian.org>
naoNao89
added a commit
to naoNao89/coreutils
that referenced
this pull request
Jan 1, 2026
…s#9136) * feat(dd): add comprehensive benchmark suite for O_DIRECT optimization - Create dd's first benchmark suite using divan framework - Benchmark various block sizes (4K, 8K, 64K, 1M) to measure performance - Test different dd scenarios: default, partial copy, skip, seek operations - Measure impact of separate input/output block sizes - All benchmarks use status=none to avoid output noise - Benchmarks verify the O_DIRECT buffer alignment optimization - Follows existing uutils benchmark patterns and conventions * bench(dd): increase dataset sizes for consistent timing Increase benchmark dataset sizes to achieve consistent 100-300ms timing: - dd_copy_default: 16 -> 32 MB - dd_copy_4k_blocks: 16 -> 24 MB - dd_copy_64k_blocks: 16 -> 64 MB - dd_copy_1m_blocks: 16 -> 128 MB - dd_copy_separate_blocks: 16 -> 48 MB - dd_copy_partial: 16 -> 32 MB - dd_copy_with_skip: 16 -> 48 MB - dd_copy_with_seek: 16 -> 48 MB - dd_copy_8k_blocks: 16 -> 32 MB This ensures stable, repeatable benchmark measurements across different systems. --------- Co-authored-by: Sylvestre Ledru <sylvestre@debian.org>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Add dd benchmark suite to validate O_DIRECT optimization theory from PR #9104.
This addresses the high priority checklist item in PR #9104. Results show 7-8x performance improvement with larger blocks (default 85.12ms vs 1M blocks 11.62ms), confirming that page-aligned buffers reduce syscall overhead
Benchmarks will track performance regression and establish baseline for O_DIRECT implementation.