Add segfault catcher.

[azteca1998]

Simplified version of #959. Does not attempt to grow the stack, just catches segfaults and returns an error.

Checklist

• Linked to Github Issue
• Unit tests added
• Integration tests added.
• This change requires new documentation.
  • Documentation has been added/updated.

[github-actions]

Benchmark results Main vs HEAD.

Command	Mean [s]	Min [s]	Max [s]	Relative
base dict_insert.cairo (JIT)	3.834 ± 0.021	3.805	3.876	1.00
base dict_insert.cairo (AOT)	3.970 ± 0.019	3.944	3.996	1.04 ± 0.01
Command	Mean [s]	Min [s]	Max [s]	Relative
:---	---:	---:	---:	---:
head dict_insert.cairo (JIT)	3.888 ± 0.035	3.850	3.967	1.00
head dict_insert.cairo (AOT)	4.020 ± 0.038	3.974	4.090	1.03 ± 0.01

Command	Mean [s]	Min [s]	Max [s]	Relative
base dict_snapshot.cairo (JIT)	3.726 ± 0.026	3.701	3.791	1.00
base dict_snapshot.cairo (AOT)	3.872 ± 0.019	3.850	3.910	1.04 ± 0.01
Command	Mean [s]	Min [s]	Max [s]	Relative
:---	---:	---:	---:	---:
head dict_snapshot.cairo (JIT)	3.745 ± 0.043	3.696	3.856	1.00
head dict_snapshot.cairo (AOT)	3.870 ± 0.013	3.850	3.895	1.03 ± 0.01

Command	Mean [s]	Min [s]	Max [s]	Relative
base factorial_2M.cairo (JIT)	4.111 ± 0.037	4.060	4.179	1.00
base factorial_2M.cairo (AOT)	4.274 ± 0.025	4.235	4.303	1.04 ± 0.01
Command	Mean [s]	Min [s]	Max [s]	Relative
:---	---:	---:	---:	---:
head factorial_2M.cairo (JIT)	4.096 ± 0.029	4.073	4.161	1.00
head factorial_2M.cairo (AOT)	4.406 ± 0.040	4.345	4.448	1.08 ± 0.01

Command	Mean [s]	Min [s]	Max [s]	Relative
base fib_2M.cairo (JIT)	3.604 ± 0.020	3.567	3.629	1.00
base fib_2M.cairo (AOT)	3.828 ± 0.015	3.805	3.848	1.06 ± 0.01
Command	Mean [s]	Min [s]	Max [s]	Relative
:---	---:	---:	---:	---:
head fib_2M.cairo (JIT)	3.750 ± 0.074	3.673	3.900	1.00
head fib_2M.cairo (AOT)	3.865 ± 0.033	3.824	3.923	1.03 ± 0.02

Command	Mean [s]	Min [s]	Max [s]	Relative
base linear_search.cairo (JIT)	3.832 ± 0.029	3.785	3.877	1.00
base linear_search.cairo (AOT)	3.968 ± 0.028	3.915	4.003	1.04 ± 0.01
Command	Mean [s]	Min [s]	Max [s]	Relative
:---	---:	---:	---:	---:
head linear_search.cairo (JIT)	3.901 ± 0.044	3.838	3.969	1.00
head linear_search.cairo (AOT)	4.036 ± 0.038	3.986	4.099	1.03 ± 0.02

Command	Mean [s]	Min [s]	Max [s]	Relative
base logistic_map.cairo (JIT)	3.904 ± 0.037	3.863	3.977	1.00
base logistic_map.cairo (AOT)	4.005 ± 0.032	3.946	4.049	1.03 ± 0.01
Command	Mean [s]	Min [s]	Max [s]	Relative
:---	---:	---:	---:	---:
head logistic_map.cairo (JIT)	4.039 ± 0.060	3.927	4.119	1.00
head logistic_map.cairo (AOT)	4.203 ± 0.042	4.141	4.282	1.04 ± 0.02

[github-actions]

Benchmarking results

Benchmark for program dict_insert

Open benchmarks

Command	Mean [s]	Min [s]	Max [s]	Relative
Cairo-vm (Rust, Cairo 1)	20.890 ± 0.107	20.754	21.079	95.07 ± 1.70
cairo-native (embedded AOT)	4.177 ± 0.041	4.100	4.240	19.01 ± 0.37
cairo-native (embedded JIT using LLVM's ORC Engine)	3.979 ± 0.022	3.946	4.017	18.11 ± 0.33
cairo-native (standalone AOT with -march=native)	0.220 ± 0.004	0.215	0.227	1.00

Benchmark for program dict_snapshot

Open benchmarks

Command	Mean [s]	Min [s]	Max [s]	Relative
Cairo-vm (Rust, Cairo 1)	5.890 ± 0.035	5.847	5.954	2909.00 ± 144.23
cairo-native (embedded AOT)	3.989 ± 0.035	3.932	4.043	1970.47 ± 98.52
cairo-native (embedded JIT using LLVM's ORC Engine)	4.018 ± 0.036	3.955	4.060	1984.39 ± 99.25
cairo-native (standalone AOT with -march=native)	0.002 ± 0.000	0.002	0.003	1.00

Benchmark for program factorial_2M

Open benchmarks

Command	Mean [s]	Min [s]	Max [s]	Relative
Cairo-vm (Rust, Cairo 1)	14.614 ± 0.094	14.507	14.779	10.73 ± 0.10
cairo-native (embedded AOT)	4.261 ± 0.026	4.228	4.303	3.13 ± 0.03
cairo-native (embedded JIT using LLVM's ORC Engine)	4.097 ± 0.033	4.063	4.153	3.01 ± 0.03
cairo-native (standalone AOT with -march=native)	1.363 ± 0.010	1.355	1.387	1.00

Benchmark for program fib_2M

Open benchmarks

Command	Mean [s]	Min [s]	Max [s]	Relative
Cairo-vm (Rust, Cairo 1)	14.516 ± 0.066	14.403	14.602	180.17 ± 2.79
cairo-native (embedded AOT)	3.785 ± 0.018	3.754	3.803	46.98 ± 0.73
cairo-native (embedded JIT using LLVM's ORC Engine)	3.578 ± 0.045	3.536	3.667	44.41 ± 0.86
cairo-native (standalone AOT with -march=native)	0.081 ± 0.001	0.080	0.088	1.00

Benchmark for program linear_search

Open benchmarks

Command	Mean [s]	Min [s]	Max [s]	Relative
Cairo-vm (Rust, Cairo 1)	6.147 ± 0.055	6.051	6.227	3031.73 ± 77.49
cairo-native (embedded AOT)	4.210 ± 0.063	4.128	4.320	2076.27 ± 58.68
cairo-native (embedded JIT using LLVM's ORC Engine)	3.947 ± 0.063	3.873	4.055	1946.54 ± 55.93
cairo-native (standalone AOT with -march=native)	0.002 ± 0.000	0.002	0.002	1.00

Benchmark for program logistic_map

Open benchmarks

Command	Mean [s]	Min [s]	Max [s]	Relative
Cairo-vm (Rust, Cairo 1)	5.869 ± 0.052	5.803	5.980	24.20 ± 0.22
cairo-native (embedded AOT)	4.160 ± 0.066	4.088	4.319	17.16 ± 0.27
cairo-native (embedded JIT using LLVM's ORC Engine)	3.980 ± 0.029	3.940	4.015	16.41 ± 0.12
cairo-native (standalone AOT with -march=native)	0.243 ± 0.000	0.242	0.243	1.00

[codecov-commenter]

Codecov Report

Attention: Patch coverage is 39.13043% with 42 lines in your changes missing coverage. Please review.

Project coverage is 80.85%. Comparing base (a6ddeee) to head (a0c3d41).
Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
src/utils/safe_runner.rs	33.33%	42 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1053      +/-   ##
==========================================
- Coverage   80.95%   80.85%   -0.10%     
==========================================
  Files         108      109       +1     
  Lines       29938    30003      +65     
==========================================
+ Hits        24237    24260      +23     
- Misses       5701     5743      +42     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[edg-l]

Is it possible to add a test that segfaults? Or too tricky?

[azteca1998]

It is technically possible, but it'd require a way to initialize the signal handler globally when testing, potentially messing up the test harness's signal handler (ex. when using nextest).

Instead, if we just initialized every (segfaulting) test then we'd end up initializing it multiple times, which could also cause problems.