Skip to content

bump rustix and xattr - #15943

Merged
cakebaker merged 2 commits into
uutils:mainfrom
oech3:xattr-rustix
Oct 29, 2026
Merged

cakebaker merged 2 commits into
uutils:mainfrom
oech3:xattr-rustix

Conversation

@oech3

@oech3 oech3 commented Oct 28, 2026 •

Copy link
Copy Markdown
Contributor

Closes #15902 Closes #15899 Closes #15945

@codspeed

codspeed Bot commented Oct 28, 2026 •

Copy link
Copy Markdown
Contributor

Merging this PR will degrade performance by 6.34%

1 warning, 1 note

⚠️ 120 benchmarks are dominated by system calls

System calls cannot be simulated consistently and are left out of the measure. The reported value understates the real cost. Switch to the Walltime instrument to measure them.

Learn more about system calls

See affected benchmarks

ℹ️ 1 benchmark has no flamegraph

The flamegraph wasn't generated: most of the time was spent in runtime internals, which would leave nothing useful to analyze.

See affected benchmarks

⚡ 1 improved benchmark
❌ 4 regressed benchmarks
✅ 396 untouched benchmarks
⏩ 70 skipped benchmarks1

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

Mode Benchmark BASE HEAD Efficiency
❌ Memory cp_recursive_deep_tree[(120, 4)] 532.6 KB 699.2 KB -23.82%
❌ CPU Simulation ls_recursive_wide_tree[(10000, 1000)] 66.3 ms 72.9 ms -9.05%
❌ CPU Simulation sort_long_line[10000] 996.7 µs 1,073.4 µs -7.15%
❌ Memory du_deep_tree[(100, 3)] 76.5 KB 82.4 KB -7.14%
⚡ CPU Simulation mv_force_overwrite 154.4 ms 128 ms +20.65%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.


Comparing oech3:xattr-rustix (0704aed) with main (27ad196)2

Open in CodSpeed

Footnotes

  1. 70 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. ↩

  2. No successful run was found on main (450dbb7) during the generation of this report, so 27ad196 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

@sylvestre

Copy link
Copy Markdown
Contributor

@codspeedbot fix this regression

@codspeed

codspeed Bot commented Oct 28, 2026 •

Copy link
Copy Markdown
Contributor

@codspeedbot fix this regression

I looked into it and there's no regression to fix. The 4 flagged benchmarks come from differences between the CI runners, not from this PR, so I didn't push any changes.

Why the report flagged them

All four regressed benchmarks (ls_recursive_wide_tree, ls_recursive_mixed_tree, du_deep_tree, du_wide_tree) had their base and head built and run in different environments:

Benchmarks Base Head
ls_* rustc 1.99.0, AMD EPYC 7763 rustc 1.99.1, AMD EPYC 9V45
du_* rustc 1.99.1, AMD EPYC 7763 rustc 1.99.0, AMD EPYC 9V74

A different compiler and a different simulated CPU cache model can move the simulated cost even when the source is the same. The code paths also rule out the bump:

  • uu_du doesn't depend on xattr at all.
  • ls_recursive_wide_tree runs ls -R without -l, so it never reads xattrs.

The ⚡ improvements in the report (mv_force_overwrite, false_consecutive_calls) are runner noise for the same reason.

Re-run on one machine

I built the uu_ls and uu_du benchmarks twice in the same sandbox with the same toolchain: once with main's Cargo.lock (rustix 1.1.5, xattr 1.6.1) and once with this PR's (rustix 1.1.6, xattr 1.7.0). Then I ran both in simulation mode:

Benchmark main lock PR lock
ls_recursive_wide_tree[(10000, 1000)] 34.96 ms 34.95 ms
ls_recursive_mixed_tree 1.19 ms 1.19 ms
du_wide_tree[(5000, 500)] 6.87 ms 6.90 ms
du_deep_tree[(100, 3)] 828.6 µs 842.7 µs

CodSpeed rates all 12 ls/du benchmarks as unchanged, with an overall impact of -0.08% (comparison). The du_deep_tree gap is within that benchmark's usual run-to-run noise.

You can acknowledge the 4 regressions on CodSpeed and merge. The benchmark jobs run on standard ubuntu-latest runners, which can differ in CPU and Rust patch version between runs, so this kind of mismatch will keep showing up. Running the benchmarks on CodSpeed Macro Runners would stop it.

Comment @codspeedbot compare <util> benchmarks on main vs this PR and I'll run the same same-machine check for any other util.

Summary
  • Warming up the performance wizard...
  • Triage reported regressions — All 4 regressed benchmarks were compared across different CPUs and different rustc (1.99.0 vs 1.99.1); du doesn't link xattr, and ls_recursive_wide_tree (no -l) never calls xattr
  • Reproduce locally base vs head — Same machine and toolchain: all 12 ls/du benchmarks unchanged
  • Report findings — Report findings
  • Handling comment — Comment handling finished successfully

Learnings updated. See what CodSpeed remembers about this repository.

Was this response helpful? React with 👍 or 👎 to help us improve CodSpeedBot.

@xtqqczze

Copy link
Copy Markdown
Collaborator

the crate bumps should be made in separate commits, or maybe we could merge #15945 separately. but first we need to fix #15937 to unblock CI

@xtqqczze

Copy link
Copy Markdown
Collaborator

@sylvestre I think #15602 will stop the benchmark variance

@cakebaker
cakebaker merged commit 984892d into uutils:main Oct 29, 2026
189 of 190 checks passed
@cakebaker

Copy link
Copy Markdown
Contributor

Thanks!

@oech3
oech3 deleted the xattr-rustix branch October 29, 2026 08:21
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

dep, test_mv_cross_device_xattr_failure_keeps_the_rest fails with rustix 1.1.6

4 participants