Use faster bincode for ipc

[bwoebi]

Description

Should bring a considerable speedup by not using json, which is comparatively slower.

Reviewer checklist

• Test coverage seems ok.
• Appropriate labels assigned.

[codecov-commenter]

Codecov Report

Merging #2544 (353aedb) into master (552b0ef) will increase coverage by 2.18%.
The diff coverage is n/a.

Additional details and impacted files

@@             Coverage Diff              @@
##             master    #2544      +/-   ##
============================================
+ Coverage     76.55%   78.73%   +2.18%     
  Complexity      267      267              
============================================
  Files           138      112      -26     
  Lines         17457    13485    -3972     
  Branches        976        0     -976     
============================================
- Hits          13364    10618    -2746     
+ Misses         3573     2867     -706     
+ Partials        520        0     -520     

Flag	Coverage Δ
appsec-extension	?
tracer-extension	78.69% <ø> (ø)
tracer-integrations	79.49% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

see 26 files with indirect coverage changes

---

Continue to review full report in Codecov by Sentry.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 552b0ef...353aedb. Read the comment docs.

[pr-commenter]

Benchmarks

Benchmark execution time: 2024-03-04 10:26:47

Comparing candidate commit 4486d06 in PR branch bob/ipc-bincode with baseline commit 552b0ef in branch master.

Found 9 performance improvements and 6 performance regressions! Performance is the same for 163 metrics, 4 unstable metrics.

scenario:ComposerTelemetryBench/benchTelemetryParsing

• 🟩 execution_time [-171.160µs; -143.640µs] or [-4.214%; -3.536%]

scenario:ComposerTelemetryBench/benchTelemetryParsing-opcache

• 🟩 execution_time [-176.700µs; -145.100µs] or [-4.328%; -3.554%]

scenario:ContextPropagationBench/benchInject64Bit

• 🟩 execution_time [-403.048ns; -172.952ns] or [-5.585%; -2.396%]

scenario:EmptyFileBench/benchEmptyFileBaseline

• 🟩 execution_time [-522.268µs; -367.712µs] or [-18.640%; -13.124%]

scenario:EmptyFileBench/benchEmptyFileBaseline-opcache

• 🟩 execution_time [-566.094µs; -391.426µs] or [-19.335%; -13.369%]

scenario:EmptyFileBench/benchEmptyFileOverhead

• 🟥 execution_time [+1.565ms; +1.629ms] or [+102.922%; +107.123%]

scenario:EmptyFileBench/benchEmptyFileOverhead-opcache

• 🟥 execution_time [+1.558ms; +1.626ms] or [+94.745%; +98.893%]

scenario:LaravelBench/benchLaravelBaseline

• 🟩 execution_time [-483.231µs; -300.689µs] or [-16.486%; -10.259%]

scenario:LaravelBench/benchLaravelBaseline-opcache

• 🟩 execution_time [-412.302µs; -251.798µs] or [-13.594%; -8.302%]

scenario:LaravelBench/benchLaravelOverhead

• 🟥 execution_time [+1.682ms; +1.791ms] or [+110.661%; +117.796%]

scenario:LaravelBench/benchLaravelOverhead-opcache

• 🟥 execution_time [+1.720ms; +1.797ms] or [+104.399%; +109.082%]

scenario:SymfonyBench/benchSymfonyBaseline

• 🟩 execution_time [-656.328µs; -565.852µs] or [-6.187%; -5.334%]

scenario:SymfonyBench/benchSymfonyBaseline-opcache

• 🟩 execution_time [-588.931µs; -500.389µs] or [-5.515%; -4.686%]

scenario:SymfonyBench/benchSymfonyOverhead

• 🟥 execution_time [+9.866ms; +9.916ms] or [+650.648%; +653.915%]

scenario:SymfonyBench/benchSymfonyOverhead-opcache

• 🟥 execution_time [+9.865ms; +9.980ms] or [+594.255%; +601.210%]

[PROFeNoM]

Running the macrobenchmarks, there doesn't seem to be any macro impact on real-world web apps.