Ethereum Core Devs Meeting 41 Agenda
lrettig opened this issue ยท 9 comments
Ethereum Core Devs Meeting 41 Agenda
Meeting Date/Time: Friday 06/29/18 at 14:00 UTC
Meeting Duration 1.5 hours
YouTube Live Stream Link
Livepeer Stream Link
Agenda
- Testing
- Client Updates
- Research Updates
- Constantinople hard fork timing and what to include (continuing conversation from last call).
a. EIP 145: Bitwise shifting instructions in EVM: pretty well-formed, but not 100% implemented or tested.
b. EIP 210: Blockhash refactoring.
c. EIP859: account abstraction.
d. EIP 1052: EXTCODEHASH Opcode.
e. EIP 1087: Net gas metering for SSTORE operations.
f. EIP 1014: Skinny CREATE2.
NOTE: Because of the conflict with ECDC we may skip Constantinople talk if there aren't enough devs to contribute to the discussion.
Please provide comments to add or correct agenda topics.
ECDC (Ethereum client dev conference) is happening this Friday in Berlin and I know a number of usual call participants will be there. The two events conflict. Is there any interest in pushing this week's call back by a week or two?
I'm afraid the all-core dev call can't answer Devcon4 questions, but maybe Hudson or someone else involved will be able to get back with details soon. Also, please recharge your battery ๐
@topealabi I like the idea behind the EIP. I would like it to have more discussion, improvements, and be more well formed before we potentially bring it up in the meeting. This EIP may be out of the scope for the core dev meeting anyway because it isn't a low-level protocol change. There can exceptions however which we can evaluate as this EIP becomes more mature.
I won't be able to attend the call. My recent updates on Turbo-Geth are in these blog posts: https://medium.com/@akhounov/recent-data-on-turbo-geth-performance-f7fb28b06a65, https://medium.com/@akhounov/turbo-geth-grant-and-disk-usage-update-92d0aa9ffd38
I will be presenting some new ideas at the ECDC tomorrow, will write a blog about them several days after that.
I am still planning to write a new Account Abstraction EIP (based on the existing one - 859, but without mandatory nonce, without CREATE2 embedded, and with transaction hash changing once tx is in the block). Also based on information here: https://ethresear.ch/t/a-recap-of-where-we-are-at-on-account-abstraction/1721
The item I would add on the agenda if I was on the call is changing network_id in Constantinople (and perhaps even change network_id every time there is a hard-fork). This is to remove some DAO-fork specific code from p2p code, as well as cleaner separation at every future fork. EIP has not been written yet.
Trying to gauge interest before writing an EIP: It might be a good idea to charge "external access" in a similar vein as EIP 1078: For example, in order to work around the "empt code leads to successful call" problem, Solidity always does an extcodesize
before each call
. Contracts basically pay twice for loading that account from disk. Gas costs could be changed so that this is only paid once per call frame or even only once per transaction. This would also make using libraries much cheaper, if you call certain functions multiple times.
ECDC (Ethereum client dev conference) is happening this Friday in Berlin and I know a number of usual call participants will be there. The two events conflict. Is there any interest in pushing this week's call back by a week or two?
What's the status now? Call [Y/n]?
The call is happening in eight minutes.