Skip to content

Conversation

@GuillaumeLagrange
Copy link
Contributor

For these times, using thread::sleep makes no sense as, at least on linux, the scheduler's timer resolution is orders of magnitude larger than nanoseconds.

For these times, using `thread::sleep` makes no sense as, at least on
linux, the scheduler's timer resolution is orders of magnitude larger
than nanoseconds.
@codspeed-hq
Copy link

codspeed-hq bot commented Nov 7, 2025

CodSpeed Performance Report

Merging #149 will degrade performances by 10%

Comparing fix/busy-sleep-time-scale-benches (e5436a5) with main (550d10f)

Summary

⚡ 5 improvements
❌ 1 regression
✅ 360 untouched

⚠️ Please fix the performance issues or acknowledge them on CodSpeed.

Benchmarks breakdown

Mode Benchmark BASE HEAD Change
WallTime iter_batched_ref_per_iteration 36 ns 40 ns -10%
WallTime from_elem[8192] 544 ns 507 ns +7.3%
WallTime from_elem_decimal[2048] 264 ns 252 ns +4.76%
WallTime sleep_100ns 62,315 ns 194 ns ×320
WallTime sleep_100us 162.2 µs 100.1 µs +62.02%
WallTime sleep_1us 63.3 µs 1.1 µs ×58

@GuillaumeLagrange GuillaumeLagrange merged commit e5436a5 into main Nov 7, 2025
58 of 61 checks passed
@GuillaumeLagrange GuillaumeLagrange deleted the fix/busy-sleep-time-scale-benches branch November 7, 2025 14:06
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.

3 participants