Skip to content

perf(levm): improve sstore #3555

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 3 commits into from
Jul 9, 2025
Merged

perf(levm): improve sstore #3555

merged 3 commits into from
Jul 9, 2025

Conversation

edg-l
Copy link
Contributor

@edg-l edg-l commented Jul 8, 2025 •
edited
Loading

Motivation

Locally the sstore bench from #3552 goes from 2x worse to a bit better than revm

Gas benchmarks improve 2x

Description

Closes #issue_number

@github-actions github-actions bot added L1 Ethereum client L2 Rollup client performance labels Jul 8, 2025
@github-actions GitHub Actions
Copy link

github-actions bot commented Jul 8, 2025

Lines of code report

Total lines added: 10
Total lines removed: 0
Total lines changed: 10

Detailed view 
+------------------------------------------------------------------------+-------+------+
| File                                                                   | Lines | Diff |
+------------------------------------------------------------------------+-------+------+
| ethrex/crates/vm/levm/src/db/gen_db.rs                                 | 244   | +8   |
+------------------------------------------------------------------------+-------+------+
| ethrex/crates/vm/levm/src/opcode_handlers/stack_memory_storage_flow.rs | 268   | +2   |
+------------------------------------------------------------------------+-------+------+

@github-actions GitHub Actions
Copy link

github-actions bot commented Jul 8, 2025

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 3.175 ± 0.016 3.158 3.211 1.00
main_levm_BubbleSort 4.397 ± 0.016 4.378 4.430 1.39 ± 0.01
pr_revm_BubbleSort 3.200 ± 0.032 3.165 3.263 1.01 ± 0.01
pr_levm_BubbleSort 4.328 ± 0.015 4.306 4.349 1.36 ± 0.01

Benchmark Results: ERC20Approval

Command Mean [s] Min [s] Max [s] Relative
main_revm_ERC20Approval 1.047 ± 0.005 1.043 1.060 1.02 ± 0.01
main_levm_ERC20Approval 1.545 ± 0.014 1.530 1.583 1.51 ± 0.02
pr_revm_ERC20Approval 1.024 ± 0.006 1.017 1.039 1.00
pr_levm_ERC20Approval 1.503 ± 0.013 1.482 1.526 1.47 ± 0.02

Benchmark Results: ERC20Mint

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_ERC20Mint 138.9 ± 1.1 137.3 140.5 1.02 ± 0.01
main_levm_ERC20Mint 261.3 ± 1.6 259.5 265.2 1.91 ± 0.01
pr_revm_ERC20Mint 136.6 ± 0.6 135.4 137.3 1.00
pr_levm_ERC20Mint 242.3 ± 2.3 239.1 246.1 1.77 ± 0.02

Benchmark Results: ERC20Transfer

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_ERC20Transfer 244.5 ± 2.4 241.5 248.6 1.02 ± 0.01
main_levm_ERC20Transfer 405.0 ± 1.4 402.5 407.8 1.69 ± 0.01
pr_revm_ERC20Transfer 239.8 ± 1.5 238.6 243.6 1.00
pr_levm_ERC20Transfer 386.5 ± 2.1 383.8 390.4 1.61 ± 0.01

Benchmark Results: Factorial

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_Factorial 234.6 ± 3.8 231.5 241.5 1.00 ± 0.02
main_levm_Factorial 442.7 ± 3.2 439.6 450.6 1.89 ± 0.03
pr_revm_Factorial 233.8 ± 3.0 232.1 240.9 1.00
pr_levm_Factorial 436.7 ± 1.2 434.9 439.0 1.87 ± 0.02

Benchmark Results: FactorialRecursive

Command Mean [s] Min [s] Max [s] Relative
main_revm_FactorialRecursive 1.622 ± 0.040 1.532 1.667 1.00
main_levm_FactorialRecursive 2.762 ± 0.032 2.695 2.803 1.70 ± 0.05
pr_revm_FactorialRecursive 1.622 ± 0.028 1.580 1.655 1.00 ± 0.03
pr_levm_FactorialRecursive 2.687 ± 0.024 2.649 2.730 1.66 ± 0.04

Benchmark Results: Fibonacci

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_Fibonacci 208.9 ± 3.8 206.5 218.5 1.00
main_levm_Fibonacci 439.9 ± 4.0 435.5 447.7 2.11 ± 0.04
pr_revm_Fibonacci 210.6 ± 5.7 206.6 221.6 1.01 ± 0.03
pr_levm_Fibonacci 429.7 ± 3.0 426.9 437.1 2.06 ± 0.04

Benchmark Results: FibonacciRecursive

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_FibonacciRecursive 886.1 ± 9.0 869.2 895.8 1.01 ± 0.02
main_levm_FibonacciRecursive 1425.8 ± 17.5 1403.5 1456.2 1.63 ± 0.03
pr_revm_FibonacciRecursive 874.1 ± 11.8 855.5 888.6 1.00
pr_levm_FibonacciRecursive 1390.7 ± 20.5 1372.8 1439.1 1.59 ± 0.03

Benchmark Results: ManyHashes

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_ManyHashes 8.6 ± 0.0 8.6 8.7 1.00
main_levm_ManyHashes 14.0 ± 0.1 13.9 14.1 1.62 ± 0.01
pr_revm_ManyHashes 8.6 ± 0.1 8.6 8.8 1.00 ± 0.01
pr_levm_ManyHashes 13.2 ± 0.1 13.0 13.4 1.53 ± 0.02

Benchmark Results: MstoreBench

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_MstoreBench 271.5 ± 5.5 266.1 282.4 1.01 ± 0.02
main_levm_MstoreBench 949.6 ± 8.2 942.3 970.5 3.54 ± 0.04
pr_revm_MstoreBench 268.6 ± 2.1 266.6 273.4 1.00
pr_levm_MstoreBench 957.8 ± 4.9 948.8 965.9 3.57 ± 0.03

Benchmark Results: Push

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_Push 302.4 ± 5.0 298.4 315.4 1.01 ± 0.02
main_levm_Push 1036.7 ± 4.1 1032.6 1047.3 3.46 ± 0.04
pr_revm_Push 300.0 ± 2.9 297.8 307.9 1.00
pr_levm_Push 1039.4 ± 3.5 1034.9 1047.6 3.46 ± 0.03

@edg-l edg-l changed the title perf(core): improve sstore perf(levm): improve sstore Jul 8, 2025
@edg-l edg-l added levm Lambda EVM implementation and removed L2 Rollup client L1 Ethereum client labels Jul 8, 2025
@edg-l edg-l removed this from ethrex_l1 Jul 8, 2025
@edg-l edg-l removed this from ethrex_l2 Jul 8, 2025
@edg-l edg-l moved this from Todo to In Progress in ethrex_performance Jul 8, 2025
@edg-l edg-l moved this from In Progress to In review in ethrex_performance Jul 9, 2025
@edg-l edg-l marked this pull request as ready for review July 9, 2025 11:13
@edg-l edg-l requested a review from a team as a code owner July 9, 2025 11:13
@github-actions GitHub Actions
Copy link

github-actions bot commented Jul 9, 2025

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 3.179 ± 0.016 3.161 3.214 1.00
main_levm_BubbleSort 4.481 ± 0.021 4.456 4.530 1.41 ± 0.01
pr_revm_BubbleSort 3.180 ± 0.006 3.171 3.186 1.00 ± 0.01
pr_levm_BubbleSort 4.460 ± 0.055 4.421 4.611 1.40 ± 0.02

Benchmark Results: ERC20Approval

Command Mean [s] Min [s] Max [s] Relative
main_revm_ERC20Approval 1.055 ± 0.043 1.032 1.176 1.02 ± 0.04
main_levm_ERC20Approval 1.553 ± 0.004 1.549 1.560 1.50 ± 0.01
pr_revm_ERC20Approval 1.035 ± 0.010 1.023 1.049 1.00
pr_levm_ERC20Approval 1.530 ± 0.008 1.520 1.547 1.48 ± 0.02

Benchmark Results: ERC20Mint

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_ERC20Mint 137.2 ± 0.5 136.1 138.0 1.00 ± 0.01
main_levm_ERC20Mint 260.8 ± 4.0 257.7 270.0 1.91 ± 0.03
pr_revm_ERC20Mint 136.8 ± 0.8 135.9 138.2 1.00
pr_levm_ERC20Mint 247.1 ± 2.2 244.8 252.2 1.81 ± 0.02

Benchmark Results: ERC20Transfer

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_ERC20Transfer 242.4 ± 2.6 238.9 246.7 1.01 ± 0.01
main_levm_ERC20Transfer 409.7 ± 1.6 407.9 412.7 1.70 ± 0.02
pr_revm_ERC20Transfer 240.8 ± 2.2 238.6 245.0 1.00
pr_levm_ERC20Transfer 394.0 ± 4.1 391.0 404.8 1.64 ± 0.02

Benchmark Results: Factorial

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_Factorial 232.2 ± 1.3 231.2 235.4 1.00 ± 0.01
main_levm_Factorial 477.1 ± 2.2 474.9 481.6 2.05 ± 0.01
pr_revm_Factorial 232.2 ± 0.5 231.1 232.9 1.00
pr_levm_Factorial 487.7 ± 22.7 474.4 531.0 2.10 ± 0.10

Benchmark Results: FactorialRecursive

Command Mean [s] Min [s] Max [s] Relative
main_revm_FactorialRecursive 1.634 ± 0.025 1.598 1.688 1.00 ± 0.02
main_levm_FactorialRecursive 2.715 ± 0.013 2.698 2.738 1.67 ± 0.03
pr_revm_FactorialRecursive 1.627 ± 0.028 1.578 1.684 1.00
pr_levm_FactorialRecursive 2.801 ± 0.056 2.764 2.915 1.72 ± 0.05

Benchmark Results: Fibonacci

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_Fibonacci 213.0 ± 15.5 205.3 255.2 1.03 ± 0.07
main_levm_Fibonacci 466.6 ± 4.0 463.5 475.9 2.26 ± 0.02
pr_revm_Fibonacci 206.7 ± 0.4 206.3 207.7 1.00
pr_levm_Fibonacci 470.9 ± 20.7 462.3 529.4 2.28 ± 0.10

Benchmark Results: FibonacciRecursive

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_FibonacciRecursive 887.1 ± 13.0 860.2 907.8 1.00 ± 0.02
main_levm_FibonacciRecursive 1425.4 ± 5.2 1416.4 1433.2 1.61 ± 0.03
pr_revm_FibonacciRecursive 885.9 ± 13.6 854.6 903.9 1.00
pr_levm_FibonacciRecursive 1481.4 ± 34.8 1444.7 1540.1 1.67 ± 0.05

Benchmark Results: ManyHashes

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_ManyHashes 8.6 ± 0.0 8.6 8.7 1.00
main_levm_ManyHashes 13.2 ± 0.1 13.0 13.4 1.53 ± 0.02
pr_revm_ManyHashes 8.7 ± 0.1 8.6 8.9 1.01 ± 0.01
pr_levm_ManyHashes 13.2 ± 0.4 12.9 14.3 1.53 ± 0.05

Benchmark Results: MstoreBench

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_MstoreBench 266.2 ± 0.8 265.0 267.7 1.00
main_levm_MstoreBench 943.0 ± 3.8 936.3 951.2 3.54 ± 0.02
pr_revm_MstoreBench 267.0 ± 1.4 265.4 270.2 1.00 ± 0.01
pr_levm_MstoreBench 947.8 ± 9.1 938.4 972.4 3.56 ± 0.04

Benchmark Results: Push

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_Push 296.8 ± 0.7 295.5 297.8 1.00
main_levm_Push 1071.2 ± 31.6 1057.4 1160.4 3.61 ± 0.11
pr_revm_Push 298.1 ± 1.0 296.3 299.8 1.00 ± 0.00
pr_levm_Push 1069.8 ± 4.1 1065.3 1077.9 3.60 ± 0.02

@github-actions GitHub Actions
Copy link

github-actions bot commented Jul 9, 2025

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 3.182 ± 0.015 3.159 3.207 1.00
main_levm_BubbleSort 4.493 ± 0.035 4.460 4.579 1.41 ± 0.01
pr_revm_BubbleSort 3.189 ± 0.010 3.174 3.208 1.00 ± 0.01
pr_levm_BubbleSort 4.485 ± 0.092 4.393 4.608 1.41 ± 0.03

Benchmark Results: ERC20Approval

Command Mean [s] Min [s] Max [s] Relative
main_revm_ERC20Approval 1.030 ± 0.006 1.021 1.037 1.00 ± 0.01
main_levm_ERC20Approval 1.554 ± 0.004 1.550 1.560 1.51 ± 0.01
pr_revm_ERC20Approval 1.027 ± 0.003 1.023 1.031 1.00
pr_levm_ERC20Approval 1.528 ± 0.005 1.523 1.537 1.49 ± 0.01

Benchmark Results: ERC20Mint

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_ERC20Mint 137.4 ± 1.6 136.0 140.4 1.01 ± 0.01
main_levm_ERC20Mint 261.6 ± 1.8 259.0 264.8 1.93 ± 0.02
pr_revm_ERC20Mint 135.8 ± 0.8 135.2 137.1 1.00
pr_levm_ERC20Mint 248.3 ± 1.5 246.1 250.3 1.83 ± 0.02

Benchmark Results: ERC20Transfer

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_ERC20Transfer 240.1 ± 2.2 237.7 244.0 1.00
main_levm_ERC20Transfer 410.4 ± 3.2 406.7 418.7 1.71 ± 0.02
pr_revm_ERC20Transfer 241.4 ± 6.2 238.2 258.5 1.01 ± 0.03
pr_levm_ERC20Transfer 394.2 ± 3.6 390.9 401.6 1.64 ± 0.02

Benchmark Results: Factorial

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_Factorial 231.8 ± 1.1 229.9 233.4 1.00
main_levm_Factorial 472.6 ± 1.5 471.0 475.7 2.04 ± 0.01
pr_revm_Factorial 233.0 ± 0.9 232.3 235.5 1.01 ± 0.01
pr_levm_Factorial 481.1 ± 17.5 473.6 530.6 2.08 ± 0.08

Benchmark Results: FactorialRecursive

Command Mean [s] Min [s] Max [s] Relative
main_revm_FactorialRecursive 1.647 ± 0.017 1.621 1.673 1.01 ± 0.02
main_levm_FactorialRecursive 2.753 ± 0.030 2.682 2.781 1.69 ± 0.03
pr_revm_FactorialRecursive 1.628 ± 0.021 1.579 1.656 1.00
pr_levm_FactorialRecursive 2.851 ± 0.074 2.764 2.972 1.75 ± 0.05

Benchmark Results: Fibonacci

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_Fibonacci 206.8 ± 0.4 206.1 207.5 1.00
main_levm_Fibonacci 462.8 ± 2.1 459.8 465.6 2.24 ± 0.01
pr_revm_Fibonacci 207.6 ± 1.3 206.6 211.2 1.00 ± 0.01
pr_levm_Fibonacci 464.2 ± 4.2 460.0 475.0 2.24 ± 0.02

Benchmark Results: FibonacciRecursive

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_FibonacciRecursive 894.4 ± 6.9 882.8 903.7 1.02 ± 0.02
main_levm_FibonacciRecursive 1419.0 ± 7.1 1408.1 1428.8 1.62 ± 0.02
pr_revm_FibonacciRecursive 873.8 ± 11.6 848.1 886.4 1.00
pr_levm_FibonacciRecursive 1481.2 ± 27.9 1455.1 1551.4 1.70 ± 0.04

Benchmark Results: ManyHashes

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_ManyHashes 8.6 ± 0.0 8.6 8.7 1.00
main_levm_ManyHashes 13.4 ± 0.2 13.2 13.7 1.55 ± 0.02
pr_revm_ManyHashes 8.7 ± 0.1 8.6 8.8 1.01 ± 0.01
pr_levm_ManyHashes 13.2 ± 0.1 13.1 13.3 1.53 ± 0.01

Benchmark Results: MstoreBench

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_MstoreBench 267.8 ± 1.7 265.9 270.5 1.00 ± 0.01
main_levm_MstoreBench 946.9 ± 3.0 943.9 953.5 3.55 ± 0.02
pr_revm_MstoreBench 267.0 ± 0.8 265.9 268.7 1.00
pr_levm_MstoreBench 946.5 ± 2.1 944.2 950.1 3.55 ± 0.01

Benchmark Results: Push

Command Mean [ms] Min [ms] Max [ms] Relative
main_revm_Push 297.5 ± 1.2 296.1 299.4 1.00 ± 0.00
main_levm_Push 1051.3 ± 2.9 1045.4 1053.9 3.54 ± 0.01
pr_revm_Push 296.9 ± 0.5 296.2 297.6 1.00
pr_levm_Push 1077.3 ± 69.1 1051.0 1273.5 3.63 ± 0.23

@github-actions GitHub Actions
Copy link

github-actions bot commented Jul 9, 2025

Benchmark for feb08df

Click to view benchmark
Test Base PR %
block payload building bench 0.2±0.00ns 0.2±0.00ns 0.00%

@github-actions GitHub Actions
Copy link

github-actions bot commented Jul 9, 2025

Benchmark for 60d97d5

Click to view benchmark
Test Base PR %
block payload building bench 0.2±0.00ns 0.2±0.01ns 0.00%

@github-actions GitHub Actions
Copy link

github-actions bot commented Jul 9, 2025 •
edited
Loading

Benchmark Block Execution Results Comparison Against Main

Command Mean [s] Min [s] Max [s] Relative
base 213.448 ± 0.710 212.321 214.597 1.00
head 214.634 ± 0.737 213.818 215.982 1.01 ± 0.00

@edg-l edg-l added this pull request to the merge queue Jul 9, 2025
Merged via the queue into main with commit d454a1b Jul 9, 2025
46 checks passed
@edg-l edg-l deleted the improve_sstore branch July 9, 2025 16:39
@github-project-automation github-project-automation bot moved this from In review to Done in ethrex_performance Jul 9, 2025
d-roak pushed a commit to 1sixtech/ethrex that referenced this pull request Jul 17, 2025
@edg-l @d-roak
perf(levm): improve sstore (lambdaclass#3555)
6827914 
**Motivation**

Locally the sstore bench from
lambdaclass#3552 goes from 2x worse to a
bit better than revm

Gas benchmarks improve 2x

<!-- Why does this pull request exist? What are its goals? -->

**Description**

<!-- A clear and concise general description of the changes this PR
introduces -->

<!-- Link to issues: Resolves lambdaclass#111, Resolves lambdaclass#222 -->

Closes #issue_number
pedrobergamini pushed a commit to pedrobergamini/ethrex that referenced this pull request Aug 24, 2025
@edg-l
perf(levm): improve sstore (lambdaclass#3555)
9b5cc90 
**Motivation**

Locally the sstore bench from
lambdaclass#3552 goes from 2x worse to a
bit better than revm

Gas benchmarks improve 2x

<!-- Why does this pull request exist? What are its goals? -->

**Description**

<!-- A clear and concise general description of the changes this PR
introduces -->

<!-- Link to issues: Resolves lambdaclass#111, Resolves lambdaclass#222 -->

Closes #issue_number
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@Arkenan Arkenan Arkenan approved these changes

@azteca1998 azteca1998 azteca1998 approved these changes

Assignees

@edg-l edg-l

Labels
levm Lambda EVM implementation performance
Projects
ethrex_performance
Status: Done
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@edg-l @Arkenan @azteca1998