Crashtracker: fix a use-after-free error

[iamluc]

Description

Reviewer checklist

• Test coverage seems ok.
• Appropriate labels assigned.

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 78.49%. Comparing base (98839cf) to head (72358cc).

Additional details and impacted files

@@             Coverage Diff              @@
##             master    #2843      +/-   ##
============================================
- Coverage     81.06%   78.49%   -2.58%     
  Complexity     2516     2516              
============================================
  Files           146      173      +27     
  Lines         14653    18682    +4029     
  Branches          0      975     +975     
============================================
+ Hits          11879    14664    +2785     
- Misses         2774     3481     +707     
- Partials          0      537     +537     

Flag	Coverage Δ
appsec-extension	69.12% <ø> (?)
tracer-extension	78.18% <ø> (ø)
tracer-php	82.27% <ø> (ø)

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

see 27 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 98839cf...72358cc. Read the comment docs.

[pr-commenter]

Benchmarks

Benchmark execution time: 2024-09-10 16:36:55

Comparing candidate commit 72358cc in PR branch luc/crashtracker-fix-use-after-free with baseline commit 98839cf in branch master.

Found 6 performance improvements and 1 performance regressions! Performance is the same for 171 metrics, 0 unstable metrics.

scenario:EmptyFileBench/benchEmptyFileBaseline-opcache

• 🟩 execution_time [-194.913µs; -62.487µs] or [-6.427%; -2.060%]

scenario:LaravelBench/benchLaravelOverhead-opcache

• 🟩 execution_time [-187.752µs; -79.208µs] or [-5.488%; -2.315%]

scenario:MessagePackSerializationBench/benchMessagePackSerialization-opcache

• 🟥 execution_time [+5.059µs; +7.361µs] or [+3.536%; +5.145%]

scenario:PDOBench/benchPDOBaseline-opcache

• 🟩 execution_time [-12.244µs; -7.576µs] or [-6.498%; -4.021%]

scenario:PDOBench/benchPDOOverhead-opcache

• 🟩 execution_time [-12.508µs; -9.592µs] or [-4.185%; -3.210%]

scenario:PDOBench/benchPDOOverheadWithDBM-opcache

• 🟩 execution_time [-13.513µs; -9.134µs] or [-4.032%; -2.726%]

scenario:WordPressBench/benchWordPressOverhead

• 🟩 execution_time [-21.842ms; -21.618ms] or [-83.027%; -82.179%]

[bwoebi]

Looks right.
Do you want to also toggle the default of the LOG_BACKTRACES setting if compiled with asan (via preprocessor directive)?

[iamluc]

Do you want to also toggle the default of the LOG_BACKTRACES setting if compiled with asan (via preprocessor directive)?

I find it easier to disable the entire signal handling. WDYT?

[bwoebi]

@iamluc Yeah, but then you can't run crashtracker tests with asan either and you'll have to disable tests via skipif too.
I'd rather just add a define in configuration.h around the default value.