perf(levm): use malachite for modexp

[edg-l]

Use malachite library to improve the performance of the modexp precompile. There are speedups up to 7-8x observed in the new benchmarks.

[github-actions]

Lines of code report

Total lines added: 12
Total lines removed: 0
Total lines changed: 12

Detailed view

+------------------------------------------+-------+------+
| File                                     | Lines | Diff |
+------------------------------------------+-------+------+
| ethrex/crates/vm/levm/src/gas_cost.rs    | 820   | +1   |
+------------------------------------------+-------+------+
| ethrex/crates/vm/levm/src/precompiles.rs | 1032  | +11  |
+------------------------------------------+-------+------+

[github-actions]

Benchmark Results Comparison

No significant difference was registered for any benchmark run.

Detailed Results

Benchmark Results: BubbleSort

Command	Mean [s]	Min [s]	Max [s]	Relative
main_revm_BubbleSort	3.134 ± 0.019	3.109	3.174	1.00
main_levm_BubbleSort	3.493 ± 0.018	3.476	3.539	1.11 ± 0.01
pr_revm_BubbleSort	3.139 ± 0.008	3.127	3.151	1.00 ± 0.01
pr_levm_BubbleSort	3.491 ± 0.029	3.466	3.552	1.11 ± 0.01

Benchmark Results: ERC20Approval

Command	Mean [s]	Min [s]	Max [s]	Relative
main_revm_ERC20Approval	1.021 ± 0.007	1.012	1.033	1.00
main_levm_ERC20Approval	1.206 ± 0.011	1.196	1.236	1.18 ± 0.01
pr_revm_ERC20Approval	1.043 ± 0.016	1.029	1.077	1.02 ± 0.02
pr_levm_ERC20Approval	1.208 ± 0.009	1.195	1.223	1.18 ± 0.01

Benchmark Results: ERC20Mint

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
main_revm_ERC20Mint	133.8 ± 0.5	133.0	134.5	1.00
main_levm_ERC20Mint	165.9 ± 1.5	163.6	168.0	1.24 ± 0.01
pr_revm_ERC20Mint	136.5 ± 0.9	135.5	138.1	1.02 ± 0.01
pr_levm_ERC20Mint	167.3 ± 2.3	165.6	172.1	1.25 ± 0.02

Benchmark Results: ERC20Transfer

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
main_revm_ERC20Transfer	237.6 ± 0.9	236.4	239.3	1.00
main_levm_ERC20Transfer	297.2 ± 2.3	293.8	299.7	1.25 ± 0.01
pr_revm_ERC20Transfer	244.7 ± 6.8	241.6	263.9	1.03 ± 0.03
pr_levm_ERC20Transfer	299.8 ± 2.1	296.6	302.6	1.26 ± 0.01

Benchmark Results: Factorial

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
main_revm_Factorial	229.6 ± 2.7	227.8	236.2	1.01 ± 0.01
main_levm_Factorial	309.2 ± 1.3	307.3	311.9	1.36 ± 0.01
pr_revm_Factorial	228.0 ± 1.2	226.9	231.0	1.00
pr_levm_Factorial	309.3 ± 1.1	307.9	312.0	1.36 ± 0.01

Benchmark Results: FactorialRecursive

Command	Mean [s]	Min [s]	Max [s]	Relative
main_revm_FactorialRecursive	1.552 ± 0.025	1.520	1.583	1.00
main_levm_FactorialRecursive	7.031 ± 0.047	6.982	7.153	4.53 ± 0.08
pr_revm_FactorialRecursive	1.555 ± 0.024	1.507	1.582	1.00 ± 0.02
pr_levm_FactorialRecursive	7.056 ± 0.025	7.015	7.084	4.55 ± 0.07

Benchmark Results: Fibonacci

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
main_revm_Fibonacci	199.6 ± 1.7	195.6	201.1	1.01 ± 0.02
main_levm_Fibonacci	300.7 ± 2.4	297.7	304.8	1.52 ± 0.03
pr_revm_Fibonacci	198.2 ± 2.9	191.7	204.0	1.00
pr_levm_Fibonacci	307.2 ± 12.8	298.1	341.4	1.55 ± 0.07

Benchmark Results: FibonacciRecursive

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
main_revm_FibonacciRecursive	827.7 ± 12.0	795.5	838.0	1.02 ± 0.02
main_levm_FibonacciRecursive	1110.9 ± 7.2	1100.2	1121.6	1.37 ± 0.03
pr_revm_FibonacciRecursive	810.9 ± 15.5	783.6	826.7	1.00
pr_levm_FibonacciRecursive	1116.5 ± 4.5	1111.1	1124.6	1.38 ± 0.03

Benchmark Results: ManyHashes

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
main_revm_ManyHashes	8.7 ± 0.6	8.4	10.5	1.01 ± 0.07
main_levm_ManyHashes	10.2 ± 0.1	10.1	10.4	1.18 ± 0.02
pr_revm_ManyHashes	8.6 ± 0.1	8.5	8.8	1.00
pr_levm_ManyHashes	10.2 ± 0.1	10.1	10.5	1.18 ± 0.02

Benchmark Results: MstoreBench

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
main_revm_MstoreBench	264.4 ± 0.7	263.4	265.5	1.00
main_levm_MstoreBench	794.6 ± 4.6	790.4	806.6	3.00 ± 0.02
pr_revm_MstoreBench	265.4 ± 1.6	263.8	268.4	1.00 ± 0.01
pr_levm_MstoreBench	791.5 ± 2.1	789.1	795.8	2.99 ± 0.01

Benchmark Results: Push

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
main_revm_Push	286.5 ± 0.8	285.5	288.2	1.00 ± 0.00
main_levm_Push	923.2 ± 3.7	920.0	932.5	3.23 ± 0.01
pr_revm_Push	285.7 ± 0.6	284.9	286.7	1.00
pr_levm_Push	923.1 ± 4.9	917.8	931.7	3.23 ± 0.02

Benchmark Results: SstoreBench_no_opt

Command	Mean [ms]	Min [ms]	Max [ms]	Relative
main_revm_SstoreBench_no_opt	157.1 ± 1.0	155.9	158.4	1.94 ± 0.03
main_levm_SstoreBench_no_opt	81.4 ± 2.7	79.6	88.8	1.01 ± 0.04
pr_revm_SstoreBench_no_opt	157.0 ± 1.2	155.4	158.4	1.94 ± 0.03
pr_levm_SstoreBench_no_opt	80.8 ± 0.9	79.5	82.3	1.00

[Arkenan]

LGTM great PR

[JereSalo]

The "EVM" aspect of it is perfectly fine. Maybe someone with more math knowledge would be able to give a deeper review in terms of the precompile itself.

[github-actions]

Benchmark Block Execution Results Comparison Against Main

Command	Mean [s]	Min [s]	Max [s]	Relative
base	162.004 ± 0.550	161.178	162.567	1.00 ± 0.00
head	161.724 ± 0.517	160.726	162.262	1.00