fix: emit all allocas in the init block instead of the current block

[TomerStarkware]

Summary

Several places emitted llvm.alloca into the libfunc's own block rather than the function's init (pre-entry) block. When that block sits inside a loop, the alloca re-executes on every iteration and the stack grows without bound.

All allocas now go through the init block, which runs exactly once per function call:

• felt252_dict_squash: range-check and gas pointers.
• squashed_dict into_entries: result array slot.
• qm31 binary ops: lhs/rhs operand slots.
• RuntimeBindingsMeta::libfunc_qm31_bin_op: result slot. Now takes the LibfuncHelper instead of a bare &Module (same as dict_get); callers already passed the helper via deref.
• trace_dump::build_state_snapshot (with-trace-dump feature): per-variable snapshot slots emitted at every statement. Takes an explicit init block; both call sites in compiler.rs pass the pre-entry block.

Test plan

• cargo check with and without with-trace-dump
• cargo clippy --all-targets clean, cargo fmt applied
• cargo test --lib -- qm31 felt252_dict squashed_dict (41 passed)

🤖 Generated with Claude Code

---

This change is 

[orizi]

@orizi reviewed 6 files and all commit messages, and made 1 comment.
Reviewable status: complete! all files reviewed, all discussions resolved (waiting on TomerStarkware).

[github-actions]

Benchmark results Main vs HEAD.

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base dict_insert.cairo (JIT)	1.491 ± 0.012	1.473	1.513	1.02 ± 0.01
base dict_insert.cairo (AOT)	1.466 ± 0.009	1.452	1.478	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head dict_insert.cairo (JIT)	1.510 ± 0.060	1.457	1.675	1.02 ± 0.04
head dict_insert.cairo (AOT)	1.485 ± 0.028	1.460	1.558	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base dict_snapshot.cairo (JIT)	1.334 ± 0.012	1.317	1.351	1.02 ± 0.01
base dict_snapshot.cairo (AOT)	1.313 ± 0.013	1.298	1.337	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head dict_snapshot.cairo (JIT)	1.324 ± 0.010	1.306	1.344	1.02 ± 0.01
head dict_snapshot.cairo (AOT)	1.293 ± 0.006	1.281	1.301	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base factorial_2M.cairo (JIT)	1.371 ± 0.013	1.355	1.397	1.01 ± 0.02
base factorial_2M.cairo (AOT)	1.355 ± 0.019	1.334	1.398	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head factorial_2M.cairo (JIT)	1.373 ± 0.008	1.360	1.386	1.00 ± 0.01
head factorial_2M.cairo (AOT)	1.368 ± 0.013	1.351	1.394	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base fib_2M.cairo (JIT)	1.324 ± 0.012	1.310	1.355	1.02 ± 0.01
base fib_2M.cairo (AOT)	1.298 ± 0.007	1.282	1.306	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head fib_2M.cairo (JIT)	1.351 ± 0.034	1.323	1.414	1.04 ± 0.03
head fib_2M.cairo (AOT)	1.304 ± 0.004	1.299	1.308	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base linear_search.cairo (JIT)	1.332 ± 0.007	1.320	1.348	1.01 ± 0.01
base linear_search.cairo (AOT)	1.314 ± 0.007	1.306	1.326	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head linear_search.cairo (JIT)	1.338 ± 0.005	1.328	1.344	1.01 ± 0.01
head linear_search.cairo (AOT)	1.319 ± 0.009	1.305	1.335	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base logistic_map.cairo (JIT)	1.320 ± 0.008	1.300	1.328	1.02 ± 0.01
base logistic_map.cairo (AOT)	1.297 ± 0.007	1.287	1.309	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head logistic_map.cairo (JIT)	1.333 ± 0.005	1.327	1.343	1.02 ± 0.01
head logistic_map.cairo (AOT)	1.307 ± 0.006	1.298	1.319	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)	11.259 ± 0.033	11.222	11.311	6.27 ± 0.11
cairo-native (embedded AOT)	1.796 ± 0.030	1.767	1.871	1.00
cairo-native (embedded JIT using LLVM's ORC Engine)	1.796 ± 0.013	1.777	1.821	1.00 ± 0.02

Benchmark for program dict_snapshot

Open benchmarks

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
Cairo-vm (Rust, Cairo 1)	514.2 ± 4.9	508.0	525.3	1.00
cairo-native (embedded AOT)	1618.5 ± 20.7	1594.8	1655.5	3.15 ± 0.05
cairo-native (embedded JIT using LLVM's ORC Engine)	1632.4 ± 15.7	1602.9	1658.6	3.17 ± 0.04

Benchmark for program factorial_2M

Open benchmarks

Command	Mean [s]	Min [s]	Max [s]	Relative
Cairo-vm (Rust, Cairo 1)	5.036 ± 0.038	4.993	5.128	2.99 ± 0.03
cairo-native (embedded AOT)	1.683 ± 0.008	1.664	1.695	1.00
cairo-native (embedded JIT using LLVM's ORC Engine)	1.686 ± 0.017	1.658	1.712	1.00 ± 0.01

Benchmark for program fib_2M

Open benchmarks

Command	Mean [s]	Min [s]	Max [s]	Relative
Cairo-vm (Rust, Cairo 1)	4.952 ± 0.023	4.912	4.987	3.06 ± 0.02
cairo-native (embedded AOT)	1.616 ± 0.007	1.604	1.625	1.00
cairo-native (embedded JIT using LLVM's ORC Engine)	1.642 ± 0.016	1.616	1.671	1.02 ± 0.01

Benchmark for program linear_search

Open benchmarks

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
Cairo-vm (Rust, Cairo 1)	559.2 ± 7.2	552.9	575.4	1.00
cairo-native (embedded AOT)	1641.0 ± 11.7	1627.9	1666.9	2.93 ± 0.04
cairo-native (embedded JIT using LLVM's ORC Engine)	1666.5 ± 9.1	1647.8	1675.4	2.98 ± 0.04

Benchmark for program logistic_map

Open benchmarks

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
Cairo-vm (Rust, Cairo 1)	469.6 ± 1.7	466.4	472.6	1.00
cairo-native (embedded AOT)	1618.2 ± 19.7	1602.5	1665.6	3.45 ± 0.04
cairo-native (embedded JIT using LLVM's ORC Engine)	1643.3 ± 13.7	1621.4	1669.4	3.50 ± 0.03