Regarding validator's in-mapping key calculation

[dantaik]

Currently, the key is calculated as bytes32 key = keccak256(abi.encodePacked(bytes16(0), pubKey)). I believe we can simplify this by using bytes32 key = keccak256(pubKey) directly, as it removes unnecessary encoding steps while still providing a unique hash for the public key.

[dantaik]

I'm assuming the above code is irrelevant to the following code in EIP4788.sol.

 // Validator's BLS public key is verified against the hash tree root within Validator chunks
bytes32 pubKeyHashTreeRoot = sha256(abi.encodePacked(validatorBLSPubKey, bytes16(0)));

[AnshuJalan]

Yes, this is unrelated. I think the redundancy stems from a previous commit where we did need to add the padding for cryptographic reasons. It is no longer required and can be removed! Thanks for spotting.

[AnshuJalan]

Sorry! We do need it. I mixed it up with the hash tree root. So, the reason is that in PreconfRegistry we hash the compressed public key that is of the type uint256[2], with first 16 bytes in first word being 0. However, in PreconfTaskManager, we simply get the 48 bytes long public key, so we need to add the 16 bytes of 0-padding before hashing.

[dantaik]

Thank you for the clarification. Do you think this minor improvement makes sense?

https://github.com/NethermindEth/Taiko-Preconf-AVS/pull/174/files

[AnshuJalan]

Having a function is helpful! but the 48 bytes length and validation checks are redundant since that is implicitly being done in the merkleization + proof verification step.