[TomerStarkware]

The macOS test job rebuilt every dependency from scratch on each run, making it the slowest job in CI (~64m). This adds the same pinned Swatinem/rust-cache step the Linux test job already uses, placed right after the toolchain install.

Splitting test-cairo into its own job is left for a follow-up so the cache win can be measured on its own first.

🤖 Generated with Claude Code

---

This change is 

[github-actions]

Benchmark results Main vs HEAD.

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base dict_insert.cairo (JIT)	1.612 ± 0.006	1.602	1.619	1.00
base dict_insert.cairo (AOT)	1.732 ± 0.297	1.603	2.577	1.07 ± 0.18

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head dict_insert.cairo (JIT)	1.212 ± 0.016	1.196	1.255	1.00
head dict_insert.cairo (AOT)	1.214 ± 0.012	1.197	1.229	1.00 ± 0.02

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base dict_snapshot.cairo (JIT)	1.355 ± 0.013	1.333	1.384	1.01 ± 0.03
base dict_snapshot.cairo (AOT)	1.341 ± 0.037	1.314	1.432	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head dict_snapshot.cairo (JIT)	1.016 ± 0.010	1.004	1.031	1.05 ± 0.01
head dict_snapshot.cairo (AOT)	0.972 ± 0.006	0.965	0.983	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base factorial_2M.cairo (JIT)	1.386 ± 0.004	1.381	1.392	1.00
base factorial_2M.cairo (AOT)	1.393 ± 0.011	1.378	1.414	1.01 ± 0.01

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head factorial_2M.cairo (JIT)	1.013 ± 0.005	1.006	1.021	1.00
head factorial_2M.cairo (AOT)	1.014 ± 0.016	1.001	1.046	1.00 ± 0.02

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base fib_2M.cairo (JIT)	1.365 ± 0.019	1.340	1.405	1.00
base fib_2M.cairo (AOT)	1.378 ± 0.030	1.340	1.440	1.01 ± 0.03

Head

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
head fib_2M.cairo (JIT)	989.1 ± 7.3	981.1	1001.6	1.00 ± 0.02
head fib_2M.cairo (AOT)	984.5 ± 19.9	963.1	1025.0	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base linear_search.cairo (JIT)	1.406 ± 0.033	1.364	1.454	1.03 ± 0.03
base linear_search.cairo (AOT)	1.363 ± 0.023	1.340	1.407	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head linear_search.cairo (JIT)	1.005 ± 0.013	0.996	1.031	1.01 ± 0.03
head linear_search.cairo (AOT)	0.998 ± 0.023	0.985	1.062	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base logistic_map.cairo (JIT)	1.354 ± 0.009	1.342	1.369	1.01 ± 0.01
base logistic_map.cairo (AOT)	1.337 ± 0.015	1.319	1.372	1.00

Head

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
head logistic_map.cairo (JIT)	997.5 ± 7.6	985.7	1015.7	1.02 ± 0.03
head logistic_map.cairo (AOT)	978.1 ± 24.3	961.8	1046.1	1.00

[github-actions]

Benchmarking results

Benchmark for program dict_insert

Open benchmarks

Command	Mean [s]	Min [s]	Max [s]	Relative
Cairo-vm (Rust, Cairo 1)	15.155 ± 0.155	14.961	15.449	7.52 ± 0.09
cairo-native (embedded AOT)	2.015 ± 0.011	2.006	2.037	1.00
cairo-native (embedded JIT using LLVM's ORC Engine)	2.031 ± 0.008	2.020	2.050	1.01 ± 0.01

Benchmark for program dict_snapshot

Open benchmarks

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
Cairo-vm (Rust, Cairo 1)	565.8 ± 5.9	555.8	573.7	1.00
cairo-native (embedded AOT)	1686.7 ± 19.1	1664.2	1722.1	2.98 ± 0.05
cairo-native (embedded JIT using LLVM's ORC Engine)	1718.7 ± 13.2	1687.6	1730.7	3.04 ± 0.04

Benchmark for program factorial_2M

Open benchmarks

Command	Mean [s]	Min [s]	Max [s]	Relative
Cairo-vm (Rust, Cairo 1)	7.142 ± 0.096	7.073	7.386	4.09 ± 0.06
cairo-native (embedded AOT)	1.745 ± 0.011	1.727	1.767	1.00
cairo-native (embedded JIT using LLVM's ORC Engine)	1.753 ± 0.008	1.741	1.765	1.00 ± 0.01

Benchmark for program fib_2M

Open benchmarks

Command	Mean [s]	Min [s]	Max [s]	Relative
Cairo-vm (Rust, Cairo 1)	7.082 ± 0.094	7.010	7.281	4.12 ± 0.09
cairo-native (embedded AOT)	1.719 ± 0.029	1.685	1.766	1.00
cairo-native (embedded JIT using LLVM's ORC Engine)	1.749 ± 0.019	1.717	1.776	1.02 ± 0.02

Benchmark for program linear_search

Open benchmarks

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
Cairo-vm (Rust, Cairo 1)	593.2 ± 1.9	590.2	595.4	1.00
cairo-native (embedded AOT)	1696.9 ± 7.5	1684.7	1706.2	2.86 ± 0.02
cairo-native (embedded JIT using LLVM's ORC Engine)	1739.3 ± 13.3	1723.6	1765.9	2.93 ± 0.02

Benchmark for program logistic_map

Open benchmarks

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
Cairo-vm (Rust, Cairo 1)	509.5 ± 1.4	507.5	512.9	1.00
cairo-native (embedded AOT)	1676.7 ± 11.1	1659.3	1696.3	3.29 ± 0.02
cairo-native (embedded JIT using LLVM's ORC Engine)	1703.9 ± 13.2	1686.3	1725.8	3.34 ± 0.03

[orizi]

@orizi reviewed 1 file and all commit messages, and made 1 comment.
Reviewable status: complete! all files reviewed, all discussions resolved (waiting on TomerStarkware).