Use Montgomery representation for Felts

[FrancoGiachetta]

Use Montgomery representation for Felts

Closes #NA

Introduces Breaking Changes?

Yes/No.

• Created PR in sequencer
• Created PR in starknet-replay
• Updated the starknet-blocks.yml workflow to use these PRs.

These PRs should be merged after this one right away, in that order.

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.

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base dict_insert.cairo (JIT)	1.969 ± 0.013	1.947	1.985	1.03 ± 0.01
base dict_insert.cairo (AOT)	1.904 ± 0.016	1.873	1.932	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head dict_insert.cairo (JIT)	2.034 ± 0.047	1.984	2.119	1.04 ± 0.03
head dict_insert.cairo (AOT)	1.947 ± 0.041	1.900	2.051	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base dict_snapshot.cairo (JIT)	1.706 ± 0.018	1.678	1.732	1.07 ± 0.02
base dict_snapshot.cairo (AOT)	1.600 ± 0.018	1.566	1.625	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head dict_snapshot.cairo (JIT)	1.764 ± 0.041	1.728	1.866	1.04 ± 0.03
head dict_snapshot.cairo (AOT)	1.701 ± 0.031	1.637	1.740	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base factorial_2M.cairo (JIT)	2.108 ± 0.020	2.083	2.152	1.02 ± 0.01
base factorial_2M.cairo (AOT)	2.061 ± 0.017	2.035	2.099	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head factorial_2M.cairo (JIT)	1.700 ± 0.034	1.656	1.769	1.03 ± 0.03
head factorial_2M.cairo (AOT)	1.651 ± 0.042	1.596	1.739	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base fib_2M.cairo (JIT)	1.652 ± 0.012	1.638	1.671	1.04 ± 0.01
base fib_2M.cairo (AOT)	1.587 ± 0.018	1.563	1.611	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head fib_2M.cairo (JIT)	1.705 ± 0.038	1.665	1.794	1.04 ± 0.02
head fib_2M.cairo (AOT)	1.633 ± 0.014	1.614	1.662	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base linear_search.cairo (JIT)	1.760 ± 0.016	1.738	1.788	1.08 ± 0.02
base linear_search.cairo (AOT)	1.630 ± 0.018	1.598	1.651	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head linear_search.cairo (JIT)	1.798 ± 0.011	1.782	1.816	1.06 ± 0.02
head linear_search.cairo (AOT)	1.699 ± 0.022	1.660	1.731	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base logistic_map.cairo (JIT)	1.926 ± 0.019	1.896	1.961	1.09 ± 0.01
base logistic_map.cairo (AOT)	1.762 ± 0.010	1.747	1.778	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head logistic_map.cairo (JIT)	1.907 ± 0.011	1.886	1.918	1.11 ± 0.02
head logistic_map.cairo (AOT)	1.724 ± 0.029	1.671	1.777	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)	11.010 ± 0.092	10.910	11.229	5.27 ± 0.09
cairo-native (embedded AOT)	2.089 ± 0.030	2.055	2.151	1.00
cairo-native (embedded JIT using LLVM's ORC Engine)	2.157 ± 0.022	2.127	2.199	1.03 ± 0.02

Benchmark for program dict_snapshot

Open benchmarks

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
Cairo-vm (Rust, Cairo 1)	596.8 ± 5.4	589.7	605.9	1.00
cairo-native (embedded AOT)	1756.9 ± 27.1	1705.5	1799.2	2.94 ± 0.05
cairo-native (embedded JIT using LLVM's ORC Engine)	1862.6 ± 16.9	1827.3	1884.5	3.12 ± 0.04

Benchmark for program factorial_2M

Open benchmarks

Command	Mean [s]	Min [s]	Max [s]	Relative
Cairo-vm (Rust, Cairo 1)	4.867 ± 0.014	4.851	4.888	2.75 ± 0.05
cairo-native (embedded AOT)	1.771 ± 0.034	1.709	1.819	1.00
cairo-native (embedded JIT using LLVM's ORC Engine)	1.835 ± 0.037	1.780	1.902	1.04 ± 0.03

Benchmark for program factorial_2M_inv

Open benchmarks

Command	Mean [s]	Min [s]	Max [s]	Relative
Cairo-vm (Rust, Cairo 1)	12.934 ± 0.121	12.738	13.091	2.04 ± 0.02
cairo-native (embedded AOT)	6.342 ± 0.023	6.299	6.389	1.00
cairo-native (embedded JIT using LLVM's ORC Engine)	6.373 ± 0.027	6.324	6.408	1.00 ± 0.01

Benchmark for program fib_2M

Open benchmarks

Command	Mean [s]	Min [s]	Max [s]	Relative
Cairo-vm (Rust, Cairo 1)	4.788 ± 0.027	4.746	4.825	2.72 ± 0.04
cairo-native (embedded AOT)	1.760 ± 0.021	1.718	1.788	1.00
cairo-native (embedded JIT using LLVM's ORC Engine)	1.832 ± 0.030	1.773	1.870	1.04 ± 0.02

Benchmark for program linear_search

Open benchmarks

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
Cairo-vm (Rust, Cairo 1)	642.6 ± 6.2	634.5	654.5	1.00
cairo-native (embedded AOT)	1786.9 ± 29.9	1750.6	1845.7	2.78 ± 0.05
cairo-native (embedded JIT using LLVM's ORC Engine)	1949.5 ± 25.3	1890.8	1978.8	3.03 ± 0.05

Benchmark for program logistic_map

Open benchmarks

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
Cairo-vm (Rust, Cairo 1)	538.5 ± 8.2	521.4	547.5	1.00
cairo-native (embedded AOT)	1800.3 ± 20.2	1770.1	1823.5	3.34 ± 0.06
cairo-native (embedded JIT using LLVM's ORC Engine)	1947.6 ± 33.0	1913.9	2010.2	3.62 ± 0.08

[github-actions]

✅ Code is now correctly formatted.

[codecov-commenter]

Codecov Report

❌ Patch coverage is 78.31169% with 167 lines in your changes missing coverage. Please review.
✅ Project coverage is 79.55%. Comparing base (5df2d2c) to head (23e2aad).

Files with missing lines	Patch %	Lines
src/starknet.rs	0.84%	117 Missing ⚠️
src/libfuncs/starknet.rs	50.00%	26 Missing ⚠️
src/libfuncs/bounded_int.rs	48.27%	15 Missing ⚠️
src/libfuncs/cast.rs	91.30%	4 Missing ⚠️
src/libfuncs/felt252.rs	83.33%	2 Missing ⚠️
src/libfuncs/bytes31.rs	90.00%	1 Missing ⚠️
src/utils/montgomery.rs	99.70%	1 Missing ⚠️
src/values.rs	97.56%	1 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1443      +/-   ##
==========================================
+ Coverage   79.52%   79.55%   +0.02%     
==========================================
  Files         108      109       +1     
  Lines       26667    27121     +454     
==========================================
+ Hits        21208    21576     +368     
- Misses       5459     5545      +86     

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[FrancoGiachetta]

Ran a benchmark of 1000 blocks (1470000-1471000) using this commit: 685735f. The results only show a clear regression in native's performance compared to previous versions of it. This is probably related to greater use of syscalls, and other non-felt related libfuncs, which made this change to have not positive effects but rather negative ones over natives performance.
Since will this most certainly the case for the majority of the blocks, I don't see any advantages of having this feature in the code.