Call Notes
Testing
- Need to merge 1 PR
- Fuzzing updates - execution-spec-tests/src/ethereum_fuzzer_differential/__init__.py at shemnon/eof-fuzz · shemnon/execution-spec-tests · GitHub
Client Update
- No Client Updates
- Rebase off of Perctra-5
- Need at least 3 for the devnet
- Compiler support
- Reviewed the PR list
- How will small proposed changes impact?
- Not too bad.
- Hashing is more impactful
- Need to totally nail down all changes before the experimental flag gets removed. Flag may remain until mainnet is live (not even testnet, mainnet)
- Assembly syntax for EXCHANGE opcode is still undetermined (absolute byte encoding, vs nybbles, vs stack index, all off by 1 issues)
- Off by one also leaks into possible SWAP/SWAPN numbering as well (DUP/SWAP are already inconsistent)
- Frangio’s summary - Assembly Syntax for EOF Stack Instructions - HackMD
- Exact is Frangio’s recommendation
Spec
- Metadata - EIP-7834: Separate Metadata Section for EOF
- EVMONE will look into a spike
- EOFCREATE/TXCREATE Hashing
- Summary doc - Potential scenarios of updating the new contract address schemes for EOF - HackMD
- Also pre-ERC for standard contracts Comparing ethereum:master...shemnon:eof/txcreate-factories · ethereum/ERCs · GitHub
- Shipping temp check
- When there are no deal-breakers
- 2025 seems tight
- Best to tie with PeerDAS, unless there is a large differential
- So the EL and CL fork major features at the same time
- But PeerDAS/EOF not as tied as, say, withdrawals
- Devnet-1 punch list - EOFv1 final tuning · Issue #165 · ipsilon/eof · GitHub
- We need to disuss the ACCOUNTTYPE opcode - EIP-7761 - MUST/SHOULD - dealbreaker for app dev until it’s in.
- is ERC-165 a viable alternative?
- What would app devs accept?