Skip to content

common: refactor to use std lib functions and optimize bitutil operations #32512

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
Closed
wants to merge 11 commits into from
Closed

common: refactor to use std lib functions and optimize bitutil operations #32512

wants to merge 11 commits into from
+65 −56

Conversation

Sahil-4555
Copy link

@Sahil-4555 Sahil-4555 commented Aug 28, 2025 •
edited
Loading

This PR makes the following improvements:

  • Refactors code to use Go’s standard library functions instead of custom implementations for better readability and maintainability.
  • Optimizes fastANDBytes, fastORBytes, and fastXORBytes by using bit shifts, reducing unnecessary operations and improving performance & benchmark by around 50%.

fastXORBytes

goos: linux
goarch: amd64
pkg: github.com/ethereum/go-ethereum/common/bitutil
cpu: Intel(R) Core(TM) i7-8565U CPU @ 1.80GHz
             │ changes_bench.txt │         original_bench.txt          │
             │      sec/op       │   sec/op     vs base                │
FastXOR1KB-8         56.64n ± 3%   91.44n ± 3%  +61.43% (p=0.000 n=10)
FastXOR2KB-8         102.4n ± 4%   173.0n ± 1%  +69.03% (p=0.000 n=10)
FastXOR4KB-8         207.8n ± 2%   340.7n ± 2%  +64.00% (p=0.000 n=10)
geomean              106.4n        175.3n       +64.79%

             │ changes_bench.txt │         original_bench.txt          │
             │       B/op        │    B/op     vs base                 │
FastXOR1KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
FastXOR2KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
FastXOR4KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
geomean                        ²               +0.00%                ²
¹ all samples are equal
² summaries must be >0 to compute geomean

             │ changes_bench.txt │         original_bench.txt          │
             │     allocs/op     │ allocs/op   vs base                 │
FastXOR1KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
FastXOR2KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
FastXOR4KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
geomean                        ²               +0.00%                ²
¹ all samples are equal
² summaries must be >0 to compute geomean

fastANDBytes

goos: linux
goarch: amd64
pkg: github.com/ethereum/go-ethereum/common/bitutil
cpu: Intel(R) Core(TM) i7-8565U CPU @ 1.80GHz
             │ changes_bench.txt │          original_bench.txt          │
             │      sec/op       │    sec/op     vs base                │
FastAND1KB-8         58.76n ± 5%   89.81n ± 10%  +52.86% (p=0.000 n=10)
FastAND2KB-8         108.7n ± 1%   165.3n ±  3%  +52.02% (p=0.000 n=10)
FastAND4KB-8         208.9n ± 2%   324.2n ±  1%  +55.22% (p=0.000 n=10)
geomean              110.1n        168.8n        +53.36%

             │ changes_bench.txt │         original_bench.txt          │
             │       B/op        │    B/op     vs base                 │
FastAND1KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
FastAND2KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
FastAND4KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
geomean                        ²               +0.00%                ²
¹ all samples are equal
² summaries must be >0 to compute geomean

             │ changes_bench.txt │         original_bench.txt          │
             │     allocs/op     │ allocs/op   vs base                 │
FastAND1KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
FastAND2KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
FastAND4KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
geomean                        ²               +0.00%                ²
¹ all samples are equal
² summaries must be >0 to compute geomean

fastORBytes

goos: linux
goarch: amd64
pkg: github.com/ethereum/go-ethereum/common/bitutil
cpu: Intel(R) Core(TM) i7-8565U CPU @ 1.80GHz
            │ changes_bench.txt │         original_bench.txt          │
            │      sec/op       │   sec/op     vs base                │
FastOR1KB-8         55.26n ± 4%   90.00n ± 5%  +62.84% (p=0.000 n=10)
FastOR2KB-8         101.8n ± 1%   164.5n ± 4%  +61.64% (p=0.000 n=10)
FastOR4KB-8         195.9n ± 1%   321.8n ± 3%  +64.24% (p=0.000 n=10)
geomean             103.3n        168.3n       +62.90%

            │ changes_bench.txt │         original_bench.txt          │
            │       B/op        │    B/op     vs base                 │
FastOR1KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
FastOR2KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
FastOR4KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
geomean                       ²               +0.00%                ²
¹ all samples are equal
² summaries must be >0 to compute geomean

            │ changes_bench.txt │         original_bench.txt          │
            │     allocs/op     │ allocs/op   vs base                 │
FastOR1KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
FastOR2KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
FastOR4KB-8        0.000 ± 0%     0.000 ± 0%       ~ (p=1.000 n=10) ¹
geomean                       ²               +0.00%                ²
¹ all samples are equal
² summaries must be >0 to compute geomean

@Sahil-4555 Sahil-4555 changed the title Common: Refactor to use std lib functions and optimize bitutil operations common: refactor to use std lib functions and optimize bitutil operations Aug 28, 2025
AlexanderYastrebov
AlexanderYastrebov reviewed Sep 26, 2025
View reviewed changes
AlexanderYastrebov
AlexanderYastrebov reviewed Sep 26, 2025
View reviewed changes
@MariusVanDerWijden
Copy link
Member

We are not really using the bitutil, we vendored it in to use a single function afaict, but I don't think we should start optimizing it, will discuss it in triage

@fjl
Copy link
Contributor

fjl commented Oct 14, 2025

We have decided not to merge this PR.

@fjl fjl closed this Oct 14, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@fjl fjl Awaiting requested review from fjl

@zsfelfoldi zsfelfoldi Awaiting requested review from zsfelfoldi

@rjl493456442 rjl493456442 Awaiting requested review from rjl493456442

@gballet gballet Awaiting requested review from gballet

@jwasinger jwasinger Awaiting requested review from jwasinger

@MariusVanDerWijden MariusVanDerWijden Awaiting requested review from MariusVanDerWijden

@s1na s1na Awaiting requested review from s1na

@lightclient lightclient Awaiting requested review from lightclient

1 more reviewer

@AlexanderYastrebov AlexanderYastrebov AlexanderYastrebov left review comments

Reviewers whose approvals may not affect merge requirements

Assignees

No one assigned

Labels

status:triage

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

4 participants

@Sahil-4555 @MariusVanDerWijden @fjl @AlexanderYastrebov