Add Ethereum-compatible aliases for BLS12-381 #4186

Jim8y · 2025-09-22T04:27:21Z

This is requested and needed by NeoX.

Description

Adds Ethereum-compatible aliases for the BLS12-381 native contract methods (bls12_*) while
retaining the existing bls12381* surface. This keeps Neo and EVM tooling interoperable without
schema changes. Updated tests cover both naming schemes, and the expected genesis manifest now
includes the new ABI entries.

Fixes # (issue)

Type of change

New feature (non-breaking change which adds functionality)

How Has This Been Tested?

dotnet test tests/Neo.UnitTests/Neo.UnitTests.csproj --filter CryptoLib

Test Configuration:

.NET SDK 9.0.301
Default protocol settings via test harness

Checklist:

My code follows the style guidelines of this project
I have performed a self-review of my code
I have commented my code, particularly in hard-to-understand areas
I have made corresponding changes to the documentation
My changes generate no new warnings
I have added tests that prove my fix is effective or that my feature works
New and existing unit tests pass locally with my changes
Any dependent changes have been merged and published in downstream modules

src/Neo/SmartContract/Native/CryptoLib.BLS12_381.cs

vncoelho

Looks good, it can be in the next release

roman-khimov

What's preventing contracts from using the already existing names?

AnnaShaleva · 2025-09-23T09:36:46Z

src/Neo/SmartContract/Native/CryptoLib.BLS12_381.cs


+        [ContractMethod(Hardfork.HF_Gorgon, CpuFee = 1 << 19, Name = "bls12_g1add")]
+        public static InteropInterface Bls12G1Add(InteropInterface x, InteropInterface y)
+            => Bls12381Add(x, y);


Adds Ethereum-compatible aliases

CryptoLib calls won't be compatible with Ethereum anyway, so I don't think this PR can be accepted. It adds an ambiguity to the CryptoLib interface.

BTW, the same thing (if we really care about specific names which is questionable to me since it's about N3 contracts, they're different from EVM/Solidity contracts anyway) can be provided by devpack without contract modifications (and code/manifest bloat).

shargon

It only change the name?

Wi1l-B0t

It seems to be just for compatibility.

Jim8y · 2025-09-26T01:13:55Z

What's preventing contracts from using the already existing names?

hi roman @roman-khimov , this is a requested change for neox, i am not sure the detail, but i guess its related to eip standards, you or anna should know it better than me.

roman-khimov

Aliases are not needed. We need to add compatibility with https://eips.ethereum.org/EIPS/eip-2537 and the only thing that can be incompatible now is bls12381MultiExp. So there is some functional extension we need here rather than aliases.

txhsl · 2025-09-26T07:11:27Z

We need to add compatibility with https://eips.ethereum.org/EIPS/eip-2537 and the only thing that can be incompatible now is bls12381MultiExp.

Yes, the EIP-2537 has been updated several times since Feb 2024, but now it gets finalized. The history can be referred https://github.com/ethereum/EIPs/commits/master/EIPS/eip-2537.md.

erikzhang · 2025-10-20T02:19:16Z

@neo-project/ngd-shanghai Need testing.

src/Neo/SmartContract/Native/CryptoLib.BLS12_381.cs

txhsl · 2025-10-20T09:32:11Z

src/Neo/SmartContract/Native/CryptoLib.BLS12_381.cs

+                        {
+                            var scalar = ParseScalar(pair[1]);
+                            if (!scalar.IsZero)
+                                g1Accumulator += new G1Projective(g1Affine) * scalar;


We need to make sure that subgroup check is executed before any multiplication operation. This was fixed in Ethereum through ethereum/EIPs#8456.

Briefly speaking, we need:

"IsOnCurve" check after G1 point decoding and G2 point decoding, e.g. https://github.com/ethereum/go-ethereum/blob/v1.16.5/core/vm/contracts.go#L1212;

"IsInSubGroup" check before multiply and pairing computation, e.g. https://github.com/ethereum/go-ethereum/blob/v1.16.5/core/vm/contracts.go#L1005 and https://github.com/ethereum/go-ethereum/blob/v1.16.5/core/vm/contracts.go#L1173.

About the detailed implementation of these checks, please ref https://github.com/Consensys/gnark-crypto/blob/v0.19.0/ecc/bls12-381/g1.go#L193-L218 and https://github.com/Consensys/gnark-crypto/blob/v0.19.0/ecc/bls12-381/g2.go#L200-L223.

src/Neo/SmartContract/Native/CryptoLib.BLS12_381.cs

shargon · 2025-10-20T09:36:31Z

src/Neo/SmartContract/Native/CryptoLib.BLS12_381.cs

+        /// <param name="pairs">Array of [point, scalar] pairs.</param>
+        /// <returns>The accumulated point.</returns>
+        [ContractMethod(Hardfork.HF_Faun, CpuFee = 1 << 23)]
+        public static InteropInterface Bls12381MultiExp(Array pairs)


Maybe we need to check the length, a max is required or it could deny the service with 1024 pairs

Cpu should be paid in each iteration

cschuchardt88 · 2025-10-23T08:21:34Z

the right way to execute a test by filters is

 dotnet test tests/Neo.UnitTests/Neo.UnitTests.csproj -- --filter CryptoLib

superboyiii · 2025-11-07T04:09:15Z

failed_tests_20251107_120156.log
I made 10 more tests, 2 failed. This log is including failed data and created scripts. You can invoke this script to CryptoLib to get stack results in UT to compare. @Jim8y

superboyiii

My fault. My go test code has a little bug and it's fixed. Now I tried several test cases, all passed.

Wi1l-B0t · 2025-11-07T11:23:34Z

My fault. My go test code has a little bug and it's fixed. Now I tried several test cases, all passed.

Can you show your testcases.

Manual testing is not enough and testcase scripts also need to be added.

Jim8y · 2025-11-08T13:36:22Z

@txhsl

shargon · 2025-11-08T19:44:43Z

Agree with @Wi1l-B0t , the testcases should be added

shargon · 2025-11-08T19:46:04Z

src/Neo/SmartContract/Native/CryptoLib.BLS12_381.cs

+        /// <param name="pairs">Array of [point, scalar] pairs.</param>
+        /// <returns>The accumulated point.</returns>
+        [ContractMethod(Hardfork.HF_Faun, CpuFee = 1 << 23)]
+        public static InteropInterface Bls12381MultiExp(Array pairs)


Cpu should be paid in each iteration

superboyiii · 2025-11-10T04:22:45Z

My fault. My go test code has a little bug and it's fixed. Now I tried several test cases, all passed.

Can you show your testcases.

Manual testing is not enough and testcase scripts also need to be added.

I made a repo for my test tools: BLS12-381-TEST. It's including a go tool. It randomly picks G1 or G2 points and scalars and get bls12381MultiExp results due to EVM Lib: consensys/gnark-crypto/ecc/bls12-381. Then the shell will compare results with Bls12381MultiExpHelper - A C# implementation from this PR which will generate scripts and invoke CLI to simulate a real production environment. I make all random and atuomatic, test many times, so this PR is LGTM.

txhsl

The input length of Bls12381Pairing needs to be extended to pairs;
The EVM test data can be used like #4185.

Otherwise, LGTM. The encoding seems good now.

txhsl · 2025-11-10T10:36:59Z

src/Neo/SmartContract/Native/CryptoLib.BLS12_381.cs

+                        {
+                            var scalar = ParseScalar(pair[1]);
+                            if (!scalar.IsZero)
+                                g1Accumulator += new G1Projective(g1Affine) * scalar;


Similar to what I've mentioned in #4185 (comment), MultiExp() is also a choice here, ref https://github.com/ethereum/go-ethereum/blob/v1.16.7/core/vm/contracts.go#L1015. Only for performance.

superboyiii · 2025-11-13T09:30:26Z

Tested. g1add, g1mul, g2add, g2mul and pairing all work well.

Wi1l-B0t · 2025-11-13T12:55:00Z

Tested. g1add, g1mul, g2add, g2mul and pairing all work well.

Can you show your testcases and testdata ?

superboyiii · 2025-11-14T02:59:08Z

Tested. g1add, g1mul, g2add, g2mul and pairing all work well.

Can you show your testcases and testdata ?

Again: https://github.com/superboyiii/BLS12-381-TEST#
Already updated. All random pick.
Compare consensus/gnark/BLS12381 results.

shargon · 2025-11-14T08:37:56Z

Can this test be added as UT?

AnnaShaleva · 2025-11-14T08:13:34Z

src/Neo/SmartContract/Native/CryptoLib.BLS12_381.cs

                G1Projective g => new(g),
                _ => throw new ArgumentException("BLS12-381 type mismatch")
            };
+            EnsureG1PointValid(in g1a);


This change (along with the change a couple of lines below) needs a mainnet states compatibility check from @superboyiii because it changes the existing CryptoLib API.

It probably needs a hardfork guard in this case.

AnnaShaleva · 2025-11-14T08:37:19Z

src/Neo/SmartContract/Native/CryptoLib.BLS12_381.cs

        }

+        [ContractMethod(Hardfork.HF_Faun, CpuFee = 1 << 21, Name = "bls12_g1mul")]
+        public static byte[] Bls12G1Mul(byte[] input)


So far, we had bls12381Add, bls12381Mul and etc. in CryptoLib manifest. And right now we add bls12_g1add, bls12_g1mul and etc., so it's not clear from a side user PoW that these methods work with Eth-compatible serialization format. So at least the naming must be improved and unified.

The bigger question is compatibility with existing CryptoLib interface. Why can't we introduce deserialisation/serialization APIs that will deserialize/serialize from/to Eth-compatible format? This allow to remove these duplicating bls12_g1add, bls12_g1mul and etc. which makes CryptoLib's API cleaner.

This problem is similar to https://github.com/neo-project/neo/pull/4185/files/546a6d07afa3a7171891b1f761f5bb26e63528ab#r2526423615.

I agree with @AnnaShaleva, it will be easier and clearer. But we should have a clear comment on what format each method accepts

tests/Neo.UnitTests/SmartContract/Native/UT_CryptoLib.cs

superboyiii · 2025-11-20T03:33:26Z

We need add check for G2 like this, when infinity, return Identity. But seems need a hardfork for safe.

shargon

change to master please

tests/Neo.UnitTests/SmartContract/Native/UT_CryptoLib.cs

superboyiii · 2025-11-21T04:14:45Z

We need change remote branch to master, then I will make compatiblity test on mainnet to ensure no storage issue.

Wi1l-B0t · 2025-11-22T11:17:11Z

Any progress?

Co-authored-by: Owen <[email protected]>

Jim8y added the Waiting for Review label Sep 22, 2025

Add Ethereum-compatible aliases for BLS12-381

1a443fe

Jim8y force-pushed the feature/bls12-eth-aliases branch from a157ce3 to 1a443fe Compare September 22, 2025 04:32

Format codebase

67b0fbb

cschuchardt88 reviewed Sep 22, 2025

View reviewed changes

src/Neo/SmartContract/Native/CryptoLib.BLS12_381.cs Outdated Show resolved Hide resolved

vncoelho reviewed Sep 22, 2025

View reviewed changes

roman-khimov reviewed Sep 23, 2025

View reviewed changes

AnnaShaleva reviewed Sep 23, 2025

View reviewed changes

shargon reviewed Sep 24, 2025

View reviewed changes

Merge branch 'dev' into feature/bls12-eth-aliases

7f3fe50

Wi1l-B0t previously approved these changes Sep 26, 2025

View reviewed changes

roman-khimov previously requested changes Sep 26, 2025

View reviewed changes

Jim8y added 2 commits October 14, 2025 19:13

Merge branch 'dev' into feature/bls12-eth-aliases

dc9b374

Implement BLS12-381 multi exponentiation

1f82a9b

Jim8y dismissed Wi1l-B0t’s stale review via 1f82a9b October 17, 2025 02:26

Merge branch 'dev' into feature/bls12-eth-aliases

be21a84

roman-khimov reviewed Oct 20, 2025

View reviewed changes

src/Neo/SmartContract/Native/CryptoLib.BLS12_381.cs Outdated Show resolved Hide resolved

roman-khimov mentioned this pull request Oct 20, 2025

BLS12-381 multi exponentiation nspcc-dev/neo-go#4041

Open

txhsl reviewed Oct 20, 2025

View reviewed changes

Merge branch 'dev' into feature/bls12-eth-aliases

3f69ba9

shargon reviewed Oct 20, 2025

View reviewed changes

src/Neo/SmartContract/Native/CryptoLib.BLS12_381.cs Outdated Show resolved Hide resolved

Update src/Neo/SmartContract/Native/CryptoLib.BLS12_381.cs

9607bd7

shargon reviewed Oct 20, 2025

View reviewed changes

erikzhang mentioned this pull request Oct 22, 2025

Dev meeting - 10-23-2025 #4236

Closed

Jim8y added 2 commits October 23, 2025 09:58

Merge branch 'dev' into feature/bls12-eth-aliases

b6fb04b

Harden BLS12-381 multi exponentiation

5f4dbe5

superboyiii approved these changes Nov 7, 2025

View reviewed changes

Merge branch 'dev' into feature/bls12-eth-aliases

a49a6b7

Jim8y added Port-to-3.x Feature or PR must be ported to Neo 3.x branch Hardfork Core labels Nov 8, 2025

shargon requested changes Nov 8, 2025

View reviewed changes

txhsl reviewed Nov 10, 2025

View reviewed changes

Add EVM-compatible BLS12 alias entrypoints

7d468ce

AnnaShaleva reviewed Nov 14, 2025

View reviewed changes

superboyiii reviewed Nov 20, 2025

View reviewed changes

tests/Neo.UnitTests/SmartContract/Native/UT_CryptoLib.cs Show resolved Hide resolved

Jim8y added 3 commits November 20, 2025 19:05

Add EIP-2537 serialize helpers and boundary tests

df66f45

Add edge-case coverage for eth BLS serialize/deserialize

93713e8

Tweak assertion message

8dedeed

shargon reviewed Nov 20, 2025

View reviewed changes

AnnaShaleva mentioned this pull request Nov 20, 2025

[Nov. 27] Core Dev Regular Meeting Discussion #4321

Open

superboyiii reviewed Nov 21, 2025

View reviewed changes

tests/Neo.UnitTests/SmartContract/Native/UT_CryptoLib.cs Show resolved Hide resolved

shargon and others added 2 commits November 27, 2025 05:52

Update tests/Neo.UnitTests/SmartContract/Native/UT_CryptoLib.cs

f2c75e6

Co-authored-by: Owen <[email protected]>

Update tests/Neo.UnitTests/SmartContract/Native/UT_CryptoLib.cs

421bc93

Co-authored-by: Owen <[email protected]>

Add Ethereum-compatible aliases for BLS12-381 #4186

Are you sure you want to change the base?

Add Ethereum-compatible aliases for BLS12-381 #4186

Conversation

Jim8y commented Sep 22, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Description

Type of change

How Has This Been Tested?

Checklist:

Uh oh!

Uh oh!

vncoelho left a comment

Choose a reason for hiding this comment

Uh oh!

roman-khimov left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

shargon left a comment

Choose a reason for hiding this comment

Uh oh!

Wi1l-B0t left a comment

Choose a reason for hiding this comment

Uh oh!

Jim8y commented Sep 26, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

roman-khimov left a comment

Choose a reason for hiding this comment

Uh oh!

txhsl commented Sep 26, 2025

Uh oh!

erikzhang commented Oct 20, 2025

Uh oh!

Uh oh!

txhsl Oct 20, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

cschuchardt88 commented Oct 23, 2025

Uh oh!

superboyiii commented Nov 7, 2025

Uh oh!

superboyiii left a comment

Choose a reason for hiding this comment

Uh oh!

Wi1l-B0t commented Nov 7, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Jim8y commented Nov 8, 2025

Uh oh!

shargon commented Nov 8, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

superboyiii commented Nov 10, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

txhsl left a comment

Choose a reason for hiding this comment

Uh oh!

txhsl Nov 10, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

superboyiii commented Nov 13, 2025

Uh oh!

Wi1l-B0t commented Nov 13, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Jim8y commented Sep 22, 2025 •

edited

Loading

Jim8y commented Sep 26, 2025 •

edited

Loading

txhsl Oct 20, 2025 •

edited

Loading

Wi1l-B0t commented Nov 7, 2025 •

edited

Loading

superboyiii commented Nov 10, 2025 •

edited

Loading

txhsl Nov 10, 2025 •

edited

Loading

Wi1l-B0t commented Nov 13, 2025 •

edited

Loading

superboyiii commented Nov 14, 2025 •

edited

Loading

AnnaShaleva Nov 14, 2025 •

edited

Loading

AnnaShaleva Nov 14, 2025 •

edited

Loading

superboyiii commented Nov 20, 2025 •

edited

Loading

superboyiii commented Nov 21, 2025 •

edited

Loading