Update compute_el_header_block_hash for EIP-7685#3778
Conversation
|
@lightclient I would appreciate your review but I can't select you as a reviewer. |
mkalinin
left a comment
mkalinin
left a comment
There was a problem hiding this comment.
The logic looks correct to me. One nit: get_deposit_receipt_rlp (same with withdrawal requests) returns Request encoding which is not an RLP because of the type byte at the beginning, consider renaming these methods to reflect their actual semantics.
Ah yes, you're right. Thanks for pointing this out! I took this approach. |
|
Note: if #3757 gets merged, we should rename it in this PR too. |
@hwwhww So I'm guessing |
|
@jtraglia you're right! I forget to update the prefix when copy-pasting. |
The parent beacon block hash was incorrectly set to zero in ethereum#3778. Passing the state to the computation function allows correct hash computation.
3 participants
Reference: https://eips.ethereum.org/EIPS/eip-7685
This PR does the following:
deposit_receiptsandwithdrawal_requestsintorequests.0x00and0x01).parent_beacon_root.parent_beacon_rootcauses issues whenrequests_rootis not null.payload.withdrawal_requests = []inbuild_empty_execution_payload.I noticed these issues when looking into why the block hash differed from Geth.