[WIP] Reduced allocations

[devmotion]

I'm curious if and possibly to what extent benchmarks are affected when reducing allocations.

Built on top of #104.

[github-actions]

Benchmark Report for /home/runner/work/Loess.jl/Loess.jl

Job Properties

• Time of benchmarks:
  • Target: 22 Nov 2025 - 00:47
  • Baseline: 22 Nov 2025 - 00:48
• Package commits:
  • Target: aed3383
  • Baseline: c61757b
• Julia commits:
  • Target: ba1e628
  • Baseline: ba1e628
• Julia command flags:
  • Target: None
  • Baseline: -C,native,-J/opt/hostedtoolcache/julia/1.12.1/x64/lib/julia/sys.so,-g1,-O3,-e,using Pkg; Pkg.update()
• Environment variables:
  • Target: None
  • Baseline: None

Results

A ratio greater than 1.0 denotes a possible regression (marked with ❌), while a ratio less
than 1.0 denotes a possible improvement (marked with ✅). Brackets display tolerances for the benchmark estimates. Only significant results - results
that indicate possible regressions or improvements - are shown below (thus, an empty table means that all
benchmark results remained invariant between builds).

ID	time ratio	memory ratio
["random", "100"]	1.56 (5%) ❌	0.29 (1%) ✅
["random", "1000"]	1.51 (5%) ❌	1.00 (1%)
["random", "10000"]	1.39 (5%) ❌	1.61 (1%) ❌
["ties", "sine"]	1.43 (5%) ❌	1.61 (1%) ❌

Benchmark Group List

Here's a list of all the benchmark groups executed by this job:

• ["random"]
• ["ties"]

Julia versioninfo

Target

Julia Version 1.12.1
Commit ba1e628ee49 (2025-10-17 13:02 UTC)
Build Info:
  Official https://julialang.org release
Platform Info:
  OS: Linux (x86_64-linux-gnu)
      Ubuntu 24.04.3 LTS
  uname: Linux 6.11.0-1018-azure #18~24.04.1-Ubuntu SMP Sat Jun 28 04:46:03 UTC 2025 x86_64 x86_64
  CPU: AMD EPYC 7763 64-Core Processor: 
              speed         user         nice          sys         idle          irq
       #1     0 MHz        272 s          0 s         53 s        942 s          0 s
       #2     0 MHz        184 s          0 s         73 s       1021 s          0 s
       #3     0 MHz        225 s          0 s         60 s        991 s          0 s
       #4     0 MHz        668 s          0 s         49 s        533 s          0 s
  Memory: 15.620681762695312 GB (14393.28515625 MB free)
  Uptime: 138.05 sec
  Load Avg:  1.31  0.67  0.26
  WORD_SIZE: 64
  LLVM: libLLVM-18.1.7 (ORCJIT, znver3)
  GC: Built with stock GC
Threads: 1 default, 1 interactive, 1 GC (on 4 virtual cores)

Baseline

Julia Version 1.12.1
Commit ba1e628ee49 (2025-10-17 13:02 UTC)
Build Info:
  Official https://julialang.org release
Platform Info:
  OS: Linux (x86_64-linux-gnu)
      Ubuntu 24.04.3 LTS
  uname: Linux 6.11.0-1018-azure #18~24.04.1-Ubuntu SMP Sat Jun 28 04:46:03 UTC 2025 x86_64 x86_64
  CPU: AMD EPYC 7763 64-Core Processor: 
              speed         user         nice          sys         idle          irq
       #1     0 MHz        344 s          0 s         59 s       1516 s          0 s
       #2     0 MHz        241 s          0 s         79 s       1608 s          0 s
       #3     0 MHz        597 s          0 s         68 s       1262 s          0 s
       #4     0 MHz        907 s          0 s         56 s        936 s          0 s
  Memory: 15.620681762695312 GB (14084.65234375 MB free)
  Uptime: 203.19 sec
  Load Avg:  1.2  0.78  0.33
  WORD_SIZE: 64
  LLVM: libLLVM-18.1.7 (ORCJIT, znver3)
  GC: Built with stock GC
Threads: 1 default, 1 interactive, 1 GC (on 4 virtual cores)

[andreasnoack]

It doesn't seem to make a difference on the timings which aligns with the profiling.