MLIR implementation of euclidean algorithm

[DiegoCivi]

The euclidean algorithm was being built for every inverse gate in a circuit. The aim of this PR is to improve compilation times in circuit related contracts by reducing the MLIR size and this is done by declaring the euclidean algorithm only once and the just calling it.

Benchmark

The following are the compilation times it took with the no_inline attribute and with an O2 optimization level. The Before column contains the compilation times for the main branch (ref -> without the optimization done in #1348), while the After column contains the compilations times for this branch (ref -> again, without the optimization done in #1348 ). These were ran on a MacBook Air M1 with 16 GB.

Class Hash	Before (ms)	After (ms)	Improvement
0x1b5fbe104c033025dbb7fb37011781cc9344e881b4828cdaa023a80fecafde4	130730	110260	1.19x
0x3100defca27214e5f78f25e48a5b05e45899c6834cb4d34f48384c18e14dff7	125509	107107	1.17x
0x79666cdb4fc3cbcafbd74f4ea4e2855bf455c5a7c70915f5679325c54032771	123684	112580	1.10x
0xa40ecaa08eaef629d433b15f6b017461df8bea9b4e7299eb8ba3632e32b5a5	132966	118622	1.12x
0x4ffeed293927cd56686a9038a10026a2d3b9602f789d1f163c1c4ac9a822a82	177028	155397	1.14x
0x2269858a40ea0535cb373b0c981c91b907466edf6d65bcaf669760bbee0ae4d	169722	149323	1.14x
0x5ff378cb2f16804539ecb92e84f273aafbab57d450530e9fe8e87771705a673	138254	120414	1.15x
0x4edde37ca59d9dff8f4ac8945b1c4860b606abd61d74727904ad7494fccdfa9	138312	120037	1.15x

Correctness

Blocks 1560000 to 1580000 were executed and no diffs were found when comparing with the VM

Checklist

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

[github-actions]

Benchmarking results

Benchmark for program dict_insert

Open benchmarks

Command	Mean [s]	Min [s]	Max [s]	Relative
Cairo-vm (Rust, Cairo 1)	11.023 ± 0.052	10.961	11.123	4.54 ± 0.03
cairo-native (embedded AOT)	2.428 ± 0.013	2.408	2.449	1.00
cairo-native (embedded JIT using LLVM's ORC Engine)	2.519 ± 0.027	2.491	2.571	1.04 ± 0.01

Benchmark for program dict_snapshot

Open benchmarks

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
Cairo-vm (Rust, Cairo 1)	522.0 ± 4.4	515.1	530.1	1.00
cairo-native (embedded AOT)	2094.0 ± 9.8	2072.7	2105.1	4.01 ± 0.04
cairo-native (embedded JIT using LLVM's ORC Engine)	2221.0 ± 14.0	2197.1	2240.9	4.25 ± 0.05

Benchmark for program factorial_2M

Open benchmarks

Command	Mean [s]	Min [s]	Max [s]	Relative
Cairo-vm (Rust, Cairo 1)	4.792 ± 0.014	4.779	4.820	1.86 ± 0.03
cairo-native (embedded AOT)	2.576 ± 0.047	2.537	2.700	1.00
cairo-native (embedded JIT using LLVM's ORC Engine)	2.593 ± 0.014	2.571	2.615	1.01 ± 0.02

Benchmark for program fib_2M

Open benchmarks

Command	Mean [s]	Min [s]	Max [s]	Relative
Cairo-vm (Rust, Cairo 1)	4.741 ± 0.021	4.717	4.782	2.28 ± 0.02
cairo-native (embedded AOT)	2.083 ± 0.019	2.060	2.120	1.00
cairo-native (embedded JIT using LLVM's ORC Engine)	2.127 ± 0.011	2.114	2.143	1.02 ± 0.01

Benchmark for program linear_search

Open benchmarks

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
Cairo-vm (Rust, Cairo 1)	559.7 ± 7.9	546.6	570.0	1.00
cairo-native (embedded AOT)	2152.3 ± 9.3	2138.9	2166.7	3.85 ± 0.06
cairo-native (embedded JIT using LLVM's ORC Engine)	2304.7 ± 23.2	2259.3	2338.0	4.12 ± 0.07

Benchmark for program logistic_map

Open benchmarks

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
Cairo-vm (Rust, Cairo 1)	377.1 ± 6.8	371.6	390.1	1.00
cairo-native (embedded AOT)	2238.2 ± 7.0	2224.7	2244.4	5.94 ± 0.11
cairo-native (embedded JIT using LLVM's ORC Engine)	2419.1 ± 31.7	2387.0	2480.0	6.42 ± 0.14

[github-actions]

Benchmark results Main vs HEAD.

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base dict_insert.cairo (JIT)	2.638 ± 0.048	2.592	2.760	1.04 ± 0.02
base dict_insert.cairo (AOT)	2.542 ± 0.025	2.495	2.571	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head dict_insert.cairo (JIT)	2.606 ± 0.022	2.574	2.639	1.03 ± 0.01
head dict_insert.cairo (AOT)	2.523 ± 0.022	2.483	2.554	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base dict_snapshot.cairo (JIT)	2.360 ± 0.030	2.311	2.404	1.06 ± 0.02
base dict_snapshot.cairo (AOT)	2.224 ± 0.022	2.193	2.275	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head dict_snapshot.cairo (JIT)	2.328 ± 0.017	2.301	2.354	1.06 ± 0.01
head dict_snapshot.cairo (AOT)	2.194 ± 0.017	2.170	2.218	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base factorial_2M.cairo (JIT)	2.709 ± 0.028	2.667	2.748	1.02 ± 0.01
base factorial_2M.cairo (AOT)	2.661 ± 0.023	2.624	2.695	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head factorial_2M.cairo (JIT)	2.659 ± 0.030	2.611	2.711	1.01 ± 0.01
head factorial_2M.cairo (AOT)	2.629 ± 0.020	2.601	2.660	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base fib_2M.cairo (JIT)	2.253 ± 0.015	2.234	2.283	1.02 ± 0.01
base fib_2M.cairo (AOT)	2.219 ± 0.021	2.185	2.255	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head fib_2M.cairo (JIT)	2.197 ± 0.016	2.175	2.227	1.02 ± 0.01
head fib_2M.cairo (AOT)	2.148 ± 0.012	2.131	2.171	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base linear_search.cairo (JIT)	2.414 ± 0.024	2.387	2.465	1.05 ± 0.01
base linear_search.cairo (AOT)	2.291 ± 0.021	2.265	2.320	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head linear_search.cairo (JIT)	2.365 ± 0.017	2.339	2.401	1.06 ± 0.01
head linear_search.cairo (AOT)	2.241 ± 0.016	2.226	2.281	1.00

Base

Command	Mean [s]	Min [s]	Max [s]	Relative
base logistic_map.cairo (JIT)	2.528 ± 0.039	2.483	2.598	1.07 ± 0.02
base logistic_map.cairo (AOT)	2.352 ± 0.010	2.337	2.368	1.00

Head

Command	Mean [s]	Min [s]	Max [s]	Relative
head logistic_map.cairo (JIT)	2.487 ± 0.018	2.464	2.519	1.07 ± 0.01
head logistic_map.cairo (AOT)	2.318 ± 0.016	2.297	2.348	1.00

[codecov-commenter]

Codecov Report

❌ Patch coverage is 96.47887% with 5 lines in your changes missing coverage. Please review.
✅ Project coverage is 76.27%. Comparing base (bef8133) to head (a78c669).

Files with missing lines	Patch %	Lines
src/libfuncs/circuit.rs	85.71%	3 Missing ⚠️
src/metadata/runtime_bindings.rs	98.34%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1357      +/-   ##
==========================================
+ Coverage   76.22%   76.27%   +0.05%     
==========================================
  Files         111      111              
  Lines       26520    26602      +82     
==========================================
+ Hits        20214    20291      +77     
- Misses       6306     6311       +5     

☔ 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.

[JulianGCalderon]

Given that this is a generic implementation of egcd, we could rename arguments to a and b. (no need to keep the circuit-related names)

        a: Value<'c, '_>,
        b: Value<'c, '_>,

[DiegoCivi]

Agree, done in 7335dd8

[DiegoCivi]

Does it include the optimization done in #1384?

Nope, i'll change the description to clarify that

[JulianGCalderon]

Looks good!