Skip to content

refactor(array): allocate execution outputs through context - #9671

Draft
gatesn wants to merge 4 commits into
ngates/buffer-allocator-buildersfrom
ngates/buffer-allocator-execution
Draft

refactor(array): allocate execution outputs through context#9671
gatesn wants to merge 4 commits into
ngates/buffer-allocator-buildersfrom
ngates/buffer-allocator-execution

fix(array): use imported allocator in compare test

c8eae21
Select commit
Loading
Failed to load commit list.
CodSpeed HQ / CodSpeed Performance Analysis failed Sep 3, 2026 in 0s

Performance Regression: -7.72%

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

⚡ 2 improved benchmarks
❌ 7 regressed benchmarks
✅ 2163 untouched benchmarks
⏩ 206 skipped benchmarks1

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

Mode Benchmark BASE HEAD Efficiency
Simulation random_i16[0.8] 73.4 µs 93.7 µs -21.66%
Simulation decompress[u64, (4000, 1024)] 71.2 µs 85.9 µs -17.1%
Simulation take_search[(0.005, 0.05)] 20.3 µs 23.3 µs -13.11%
Simulation take_search[(0.01, 0.05)] 21 µs 24.1 µs -12.7%
WallTime words_gather_scalar_avx2[65536] 8.3 µs 9.4 µs -11.82%
Simulation take_search[(0.005, 0.1)] 26.6 µs 29.7 µs -10.48%
Simulation take_fsl_f16_force_per_index[1, 10] 53.7 µs 60 µs -10.46%
Simulation random_i16[0.95] 93.8 µs 78.2 µs +19.95%
Simulation take[duplicates/repeated/primitive/nonnull/chunks=16/indices=1000] 243.2 µs 209.3 µs +16.2%

Tip

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


Comparing ngates/buffer-allocator-execution (c8eae21) with ngates/buffer-allocator-builders (69a9f2e)

Open in CodSpeed

Footnotes

  1. 206 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.