Ethereum Core Devs Meeting 139 Agenda

[timbeiko]

Meeting Info

• May 27, 2022, 14:00 UTC
• Duration: 90 minutes
• Youtube Stream: https://youtu.be/5mMd-XHAv2Q
• Zoom: shared on #allcoredevs Discord server shortly before the call

Agenda

• Merge Updates
  • Ropsten TTD issue
  • Engine API: add getPayloadBodiesByRangeV1 to #146 execution-apis#218
  • Sepolia beacon chain #526
  • Ropsten beacon chain #525
  • Coordination-Thread: Merge Göerli execution-testnet with Prater consensus-testnet eth-clients/goerli#98
  • Difficulty Bomb Tracking
• EIP Discussions (If time permits)
  • Optimizing EIP-4844 transaction validation for mempool (using KZG proofs) EIPs#5088
  • EIP-2537 (link)
  • (Moved to next call) EIP-5027

[xinbenlv]

Can i signup to follow up EIP5081, addressing early feedback

[q9f]

Just throwing in Goerli/Prater: eth-clients/goerli#98 - FYI

Edit: Are we going to rebrand Prater to Goerli? (Or Boerli? 😆)

[timbeiko]

@xinbenlv have there been substantive changes or progress since the last call? Because of our limited time, we try and limit discussions to significant updates for EIPs.

[xinbenlv]

Got it!
The goal is when we next time come to CoreDevsMeeting, we hope to provide

1. An evaluation & analysis of whether we can allow tx to expire free, or if there is a minimum charge required to avoid DoS, seeking feedback about the validity or flaws in the evaluation.
2. A reference implementation prototype with Go-Ethereum to demonstrate the Eng cost to implement it, seeking feedback about whether there are problem with the reference implementation.

@timbeiko could you help us assess if such progress/changes are consider substantive enough for taking a timeslot in the CoreDev meeting?

To make such progress, I am anticipate we need another 2weeks, how about we push down 2 weeks to the Meeting 140?

[timbeiko]

Yeah, I think it makes sense to push it 2 weeks, so people also have a chance to review the progress prior to the call. I suggest opening an Proposal to add EIP-5081 to Shanghai issue here and sharing updates on that as well, so it doesn't get lost in closed agendas.

[djrtwo]

Want to make sure EL clients have taken a look at this engine API method to deduplicate execution payloads between CL and EL

ethereum/execution-apis#218

[qizhou]

May I have a quick discussion on EIP-5027: Remove the limit on contract code size (https://github.com/ethereum/EIPs/blob/master/EIPS/eip-5027.md)?

It should not take too much time as the major code is done and tested via geth import, but we would like to see if any further issues are found from core devs such as:

• More testing plans
• Security concerns from other layers (e.g., P2P/Snapsync/mining/etc).

This can help us to clear further hurdles of including the EIP and prepare for more comprehensive tests.

[asn-d6]

Hello! I can give a brief description (and answer questions) on a suggested EIP-4844 change which significantly reduces the transaction/block verification cost for blobs/commitments using KZG proofs (ethereum/EIPs#5088).

[timbeiko]

@qizhou has this been discussed on the R&D discord already? Looking at the EthMagicians thread it seems like this could be contentious and getting async feedback may be a good way to start.

[timbeiko]

@djrtwo @asn-d6 added to the agenda 👍

[qizhou]

@qizhou has this been discussed on the R&D discord already? Looking at the EthMagicians thread it seems like this could be contentious and getting async feedback may be a good way to start.

Sure. Thanks. I have initiated the discussion in execution-dev channel.

[timbeiko]

Thanks @qizhou! It seems there is appetite to discuss this on the call, so I've added it to the agenda :-)

[timbeiko]

@asn-d6 @qizhou @PeterCCLiu friendly heads up that given the Ropsten issue we saw today, we may not have time to get to non-merge topics during the call. If we have time at the end, we can probably have 1-2 minutes per topic, but if you'd rather move to the next call to not have to sit through 90 minutes of merge stuff before, happy to do that too.

[qizhou]

Sure! I would move EIP-5027 to the next call so that we have more time to discuss it. Thanks for letting us know ahead of the current situation.

[PeterCCLiu]

@asn-d6 @qizhou @PeterCCLiu friendly heads up that given the Ropsten issue we saw today, we may not have time to get to non-merge topics during the call. If we have time at the end, we can probably have 1-2 minutes per topic, but if you'd rather move to the next call to not have to sit through 90 minutes of merge stuff before, happy to do that too.

I'd be happy to listen anyway. If we dont have much time I can talk in the next call :)

[timbeiko]

closed in favor of #538