Skip to content

fsst decompress optimize try 2 #7001

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
joseph-isaacs wants to merge 17 commits into
base: develop
Choose a base branch
from
+1,071 −15

Optimize FSST inline view construction and escape handling

7fd4e7c
Select commit
Loading
Failed to load commit list.
Draft

fsst decompress optimize try 2 #7001

Optimize FSST inline view construction and escape handling
7fd4e7c
Select commit
Loading
Failed to load commit list.
CodSpeed HQ / CodSpeed Performance Analysis failed Mar 17, 2026 in 0s

Performance Regression: -12.23%

⚡ 48 improved benchmarks
❌ 2 regressed benchmarks
✅ 959 untouched benchmarks
🆕 40 new benchmarks
⏩ 1515 skipped benchmarks1

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

Performance Changes

Mode Benchmark BASE HEAD Efficiency
⚡ Simulation chunked_dict_fsst_canonical_into[(1000, 100, 100)] 30.1 ms 27 ms +11.34%
⚡ Simulation chunked_dict_fsst_canonical_into[(1000, 100, 10)] 3.1 ms 2.8 ms +10.78%
⚡ Simulation chunked_dict_fsst_canonical_into[(1000, 10, 100)] 28.9 ms 25.9 ms +11.82%
⚡ Simulation chunked_dict_fsst_into_canonical[(1000, 10, 100)] 29 ms 25.9 ms +11.95%
⚡ Simulation chunked_dict_fsst_canonical_into[(1000, 1000, 10)] 3.1 ms 2.8 ms +10.37%
⚡ Simulation chunked_dict_fsst_canonical_into[(1000, 1000, 100)] 30.5 ms 27.5 ms +10.84%
⚡ Simulation chunked_dict_fsst_canonical_into[(1000, 10, 10)] 3 ms 2.7 ms +11.11%
⚡ Simulation chunked_dict_fsst_into_canonical[(1000, 100, 10)] 3.1 ms 2.8 ms +10.95%
⚡ Simulation chunked_dict_fsst_into_canonical[(1000, 1000, 100)] 30.6 ms 27.6 ms +10.79%
⚡ Simulation chunked_dict_fsst_into_canonical[(1000, 100, 100)] 30.1 ms 27 ms +11.23%
⚡ Simulation chunked_dict_fsst_into_canonical[(1000, 10, 10)] 3 ms 2.7 ms +11.28%
⚡ Simulation chunked_dict_fsst_into_canonical[(1000, 1000, 10)] 3.1 ms 2.8 ms +10.27%
⚡ Simulation canonicalize_compare[(1000, 64, 4)] 373.1 µs 337.3 µs +10.63%
⚡ Simulation canonicalize_compare[(1000, 64, 8)] 415.8 µs 375.1 µs +10.86%
⚡ Simulation canonicalize_compare[(10000, 16, 8)] 1.5 ms 1.3 ms +17.19%
⚡ Simulation chunked_canonicalize_into[(10, 10000, 64, 4)] 28.1 ms 24.2 ms +16.08%
⚡ Simulation chunked_canonicalize_into[(10, 10000, 4, 4)] 7.8 ms 5.9 ms +33.11%
⚡ Simulation canonicalize_compare[(10000, 16, 4)] 1.4 ms 1.2 ms +17.23%
⚡ Simulation chunked_canonicalize_into[(1000, 100, 16, 16)] 20.5 ms 18.5 ms +11.12%
⚡ Simulation canonicalize_compare[(10000, 4, 8)] 1,173.5 µs 979.3 µs +19.83%
... ... ... ... ... ...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

Comparing claude/fsst-decompress-optimize-TylL2 (7fd4e7c) with develop (50868db)

Open in CodSpeed

Footnotes

  1. 1515 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. ↩

View more details on CodSpeed HQ