appsec: fix recv/writev calls in the face of interrupting signals

[cataphract]

Description

Reviewer checklist

• Test coverage seems ok.
• Appropriate labels assigned.

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 51.94805% with 37 lines in your changes missing coverage. Please review.

Project coverage is 72.69%. Comparing base (dc9911b) to head (41d91c4).

Files with missing lines	Patch %	Lines
appsec/src/extension/network.c	40.67%	27 Missing and 8 partials ⚠️
appsec/src/extension/commands_helpers.c	88.88%	1 Missing and 1 partial ⚠️

Additional details and impacted files

@@             Coverage Diff              @@
##             master    #3008      +/-   ##
============================================
- Coverage     72.74%   72.69%   -0.05%     
  Complexity     2750     2750              
============================================
  Files           138      138              
  Lines         15060    15049      -11     
  Branches       1026     1022       -4     
============================================
- Hits          10955    10940      -15     
- Misses         3546     3556      +10     
+ Partials        559      553       -6     

Flag	Coverage Δ
appsec-extension	67.82% <51.94%> (-0.14%)	⬇️
tracer-php	74.55% <ø> (-0.02%)	⬇️

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

Files with missing lines	Coverage Δ
appsec/src/extension/network.h	100.00% <ø> (ø)
appsec/src/extension/commands_helpers.c	62.03% <88.88%> (-1.63%)	⬇️
appsec/src/extension/network.c	43.36% <40.67%> (-0.12%)	⬇️

... and 1 file 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 dc9911b...41d91c4. Read the comment docs.

[pr-commenter]

Benchmarks [ appsec ]

Benchmark execution time: 2024-12-20 16:33:05

Comparing candidate commit 41d91c4 in PR branch glopes/recv-writev-eintr with baseline commit dc9911b in branch master.

Found 0 performance improvements and 0 performance regressions! Performance is the same for 12 metrics, 0 unstable metrics.

[estringana]

typo

[cataphract]

no, it's correct, _imsg_destroy should be called if and only if _imsg_recv returns success. I did spot a branch where it was not being called, though, so I added a new commit to use a gcc/clang extension to prevent this.