Execution Layer Meeting 179

[timbeiko]

Meeting Info

• Jan 18, 2024, 14:00-15:30 UTC
• Youtube Stream: https://youtube.com/live/FhMBatolkOM
• Zoom: shared on #allcoredevs Discord channel shortly before the call
• Ethereum Protocol Calls Calendar subscription

Agenda

• Dencun Updates
  • Goerli fork
  • Add eth_blobBaseFee; add blobs to eth_feeHistory execution-apis#486
  • Deployment of 4788 contract on testnets & mainnet
  • New CL release: release v1.4.0-beta.6 consensus-specs#3578
  • Sepolia & Holesky fork timing, see: EIP-7569
• Prague/Electra Proposals
  • EIP-3074 User Perspective, slides
  • SETCODE alternative designs
  • EL/CL EIPs:
    • EIP-7251
    • EIP-7547
  • Client team thoughts
    • reth
  • Verkle discussion next ACDE
• Besu Jan 6th Mainnet Event & Reddit AMA performance numbers

[yperbasis]

Can we please deploy the beacon roots contract on Sepolia, Holesky, mainnet.

[gakonst]

Here's the Reth team view on post-Cancun planning, keen to discuss if there's time!

[timbeiko]

@yperbasis, @marioevz is quick: sepolia, holesky 🔥

[0xfoobar]

Would love to share a quick perspective on EIP-3074 on behalf of apps and end users

[timbeiko]

@0xfoobar, let's do it! The zoom link for the call is shared in the R&D discord, in the #allcoredevs channel.

[gakonst]

@0xfoobar in order to keep scope tight, do you mind sharing some bullets / guidelines on what the app / end user view is?

I think as a community we should improve the information density and conciseness in ACD. One way to work towards that is to have more 'bulletpointed' communication, to avoid discussions going too long and w/o clear action items / next steps.

So for 3074 would love for a written pre-read with some proposed action items so we can have a productive discussion with high context and next steps, WDYT @0xfoobar?

[0xfoobar]

@gakonst Definitely. Longform rationale is here: https://0xfoobar.substack.com/p/batch-transactions-for-eoas

TLDR:

• stale approvals are a huge security risk, have already drained 100s of millions from retail. EOA batch txs fix this
• batch txs are a 'free lunch' for state growth, by removing unnecessary overhead of sequential txs can fit more actions into same state size
• batch txs eliminate unbundling risk and make the private relaying world more competitive
• EOAs are >90% of current addresses, should continue improving the default experience while working on more complex smart contract wallets

[michaelneuder]

if there is time i'd like to discuss the following two EIPs as candidates for prague! they both have EL & CL implications :-)

• EIP 7251: https://github.com/ethereum/EIPs/blob/master/EIPS/eip-7251.md
• EIP 7547: https://github.com/ethereum/EIPs/blob/master/EIPS/eip-7547.md

[wjmelements]

I want to discuss alternative designs for SETCODE to evaluate core developer preferences. While there was some resistance to the belongs-to specification, I suspect it will be favored when the rationale is explained.

• match-SSTORE specificiation, original
• belongs-to specification, proposed by yoav, current
• CODE_ADDRESS specification, proposed by Vitalik

You can preview this presentation here

[timbeiko]

@0xfoobar @wjmelements thanks for the prepared materials! I think what makes the most sense is for you both, and @michaelneuder, to go over your proposals before we hear from all the client teams (in case it shifts their views!).

That said, I want to make sure we do leave time for that at the end. So, depending on how long the Dencun conversation takes, I may timebox each specific EIP conversation. Ideally, we can do 10ish minutes each!

[hwwhww]

We will cut CL spec release v1.4.0-beta.6 immediately after ADCD today. The only consensus change is a fork choice one: ethereum/consensus-specs#3431

If some CL devs are here, I want to double check if we will use v1.4.0-beta.6 for Sepolia & the future releases for Goerli.

IMO it's okay to mix versions in Goerli since it's unlikely to cause consensus failure.

[g11tech]

If some CL devs are here, I want to double check if we will use v1.4.0-beta.6 for Sepolia & the future releases for Goerli.

Lodestar could push this change into sepolia builds, also isn't goerli future = dead?

[timbeiko]

Closed in favor of #943